Description
The report on topics like manufacturing flexibility, Organizational Performance, path analysis technique
PRODUCTION
AND OPERATIONS MANAGEMENT Vol. 5. No. 3. Fall 1996 PIinlPd in u.SA.
BUSINESS STRATEGY, MANUFACTURING FLEXIBILITY, AND ORGANIZATIONAL PERFORMANCE RELATIONSHIPS: A PATH ANALYSIS APPROACH*
YASH P. GUPTA
AND
TONI
M. SOMERS
Graduate School of Business Administration, University of Colorado at Denver, Denver, Colorado 802 17, USA Department of Finance and Business Economics, School of Business Administration, Wayne State University, Detroit, Michigan 48202, USA
It has been argued in the literature that business strategy and manufacturing flexibility independently affect the performance of an organization. However, no empirical examination of the interrelationship among thesethree constructs has been performed. In this paper, basedon a field study of 269 firms in the manufacturing industry, the identified constructs have been used to test a theoretical model using path analysis techniques. Our re+sults indicate that business strategy contributes both directly and indirectly to organizational performance. The findings provide evidenceof direct effectsof(i) businessstrategyon manufacturing flexibility and (ii) manufacturing flexibility on organizational performance. (MANUFACTURING FLEXIBILITY; BUSINESS STRATEGY; ORGANIZATIONAL PERFORMANCE; PATH MODEL)
1. Introduction
Manufacturing flexibility is one of the most difficult goals for organizations to achieve. Concomitantly, evidence suggeststhat the focus of competition in global markets is shifting from quality and service toward flexibility (Ferdows and De Meyer 1989). In Europe, for example, extensive programs are being initiated to enhance manufacturing flexibility. Theseinclude implementation of advanced manufacturing technology, decrease of cycle times through lead time and setup time reductions (Giffi, Roth, and Seal 1990). Manufacturing flexibility is a multidimensional concept (in this paper manufacturing flexibility and flexibility are used interchangeably). It ensures that the manufacturing process is both cost-efficient and -effective in that it can produce customized products without sacrificing either objective. As setup time decreases, small-batch production can be as economical as large-scale manufacturing, enabling an organization to change its competitive strategy from emphasizing economies of scale to emphasizing economies of scope(Goldhar and Jelinek 1983). Flexibility can be used both as an adaptive response to environmental uncertainty and to proactively create market uncertainties for competition (Gupta and Goyal 1989; Gerwin 1993). For example, in 1981 Honda exploited
* Received December 1993; revised December 1994; accepted April 1995. 204 lO59-1478/96/0503/204$I .25
Copyright 0 1996. Productmn and Operations Management Society
STATEGY,
FLEXIBILITY,
AND
PERFORMANCE
RELATIONSHIPS
205
the benefits of flexibility in inducing customers to expect more frequent changes from the motorcycle industry. In their battle with Yamaha for supremacy in the industry, over a period of 18 months, Honda introduced or replaced 113 models, effectively turning over its entire product line twice. Interestingly, Yamaha managed to complete only 37 changes during the same period, thus creating the consumer perception that Yamaha motorcycles were old, redundant, and unattractive compared to Honda’s. This forced Yamaha to withdraw its challenge (Stalk and Hout 1990). According to Sethi and Sethi ( 1990, p. 295 ) “manufacturing flexibility clearly has major implications for a firm’s competitive strength. This significant role of manufacturing flexibility makes it a part of the firm’s strategy.” Moreover, flexibility cannot be bought; it must be planned and managed (Gustavsson 1984). Despite the widespread acceptance of flexibility’s role in enhancing our competitive position, the management of flexibility remains poorly understood in industry (Cox 1979; Miller and Roth 1987; Tombak and De Meyer 1988; Zammuto and O’Connor 1992). Manufacturing flexibility does not receive adequate attention at the time of decision making for investment in manufacturing technology (Adler 1988), nor does it receive adequate recognition in the implementation phase (Jaikumar 1986). Slack ( 1987) in a study of 10 manufacturing companies observed that managers had a partial, rather than comprehensive, view of manufacturing ?lexibility. Managers focused on machine rather than system flexibility [using Buzacott’s ( 1982) distinction], frequently limiting themselves to a particular type of resource. While it is very tempting to think in this fashion, this strategy may result in a serious mistake. Machine level flexibility alone (e.g., versatile numerical control (NC) machines) does not adequately ensure a competitive edge and will elevate competitiveness only if the added advantage of flexibility in the management of the system (e.g., alternate routing policy) is present. In the same study, Slack ( 1987) found that managers sought to limit the need for flexibility by pursuing three broad strategies: (i) by limiting product range and discouraging frequent product modifications; (ii) by pursuing make-to-stock rather than make-to-order; and (iii) by matching market segmentation with segmentation of the production system, thus reducing the product range. On the research front, the scope of research on manufacturing flexibility has remained quite narrow. Recent literature has focused on defining types of flexibility and identifying systemsthat exhibit one or more of these (for example, seeBrowne et al. 1984;Chattejee, Cohen, Maxwell, and Miller 1984; Carter 1986; Gerwin 1987; Son and Park 1987; Brill and Mandelbaum 1989; Hyun and Ahn 1990). Few researchers have examined measurement issuesfor various types of flexibility (for example, seeMasuyama 1983; Chung and Chen 1990; Dixon 1992; Gupta and Goyal 1992; Gupta and Somers 1992). There is little agreement on how to define flexibility, how to achieve flexibility, or what are the costsand benefits of more, or less,flexibility. For example, some researchers,have viewed flexibility primarily in terms of programmable machines and capabilities for mixing models in production (Taymaz 1989). Others have viewed this only in terms of versatility of people and skills (Walton and Susman 1987). Additionally, some studiestreat flexibility and flexible manufacturing systems( FMS) as equivalent concepts. In actuality, they are not the same. FMS is one method of acquiring flexibility. Other avenues include workers with broad skills, flexible production management techniques, and the development of networks of dependable suppliers. Researchershave also implicitly or explicitly assumed that more flexibility is always better (notable exceptions are Cusumano 1988; Tombak 1988). Flexibility may actually make the players worse off (for example, see Gaimon and Singhal 1992) in some situations. On the basis of the above discussion, it is apparent that achieving manufacturing flexibility is a critical source of competitive advantage for many organizations. Upton ( 1994) while concurring with this view highlighted that:
206
CEOS
YASH P. GUPTA AND TONI M. SOMERS know this, managers know it, and shop floor operators know it. However, the exhortation heard time and time again to “go forth and be flexible” is hollow and meaningless. Managers find themselves unable to express exactly what it is that needs to be improved, without some more precise way of defining the direction of improvement. Quite often, management needs to identify multiple types of flexibility and split them up so they can be prioritized, measured and improved, each by appropriate mechanisms.
The literature on manufacturing flexibility is fragmented and does not provide answers to the basic question of manufacturing organizations: Given that flexibility is a multidimensional concept and that the need for each dimension may vary, under what level would a given flexibility be appropriate in improving organizational performance? As stated earlier, more flexibility in itself is not necessarily useful. The appropriate degree of flexibility dependson what a firm is aiming to achieve in terms of its products, behavior of its competitors, product-demand characteristics, and other factors (Suarez, Cusumano, and Fine 1992). We deal with the above question by investigating how business strategy and dimensions of manufacturing flexibility are related to organizational performance. 1.1. Motivation for Research Our analysis is motivated in part by the recent work of Get-win ( 1993) who reviewed the literature concerning the strategic aspectsof manufacturing flexibility. He established a researchagenda for the area. First, research is needed to determine the extent of which manufacturing flexibility has an impact on a company’s performance. A few investigators have reported relationships between specific types of flexibility and performance. Kekre and Srinivasan ( 1990), for example, investigated the positive and negative effects of product line breadth (in this paper we equated this to processflexibility or mix flexibility as some researchershave termed it) on firm performance. Their results provided some empirical support for the benefits of process flexibility by demonstrating that a broader product line is associatedwith significant market share benefits and increases in firms’ profitability. We recognize that at the conceptual level, equating of product line to process flexibility and then taking the relationship between product line breadth and performance as evidence of a direct relationship between flexibility and performance may be too aggressive.At the operational level, though incomplete, breadth of product line may be an acceptable measure of process flexibility. The breadth of the product line will have positive effectson market share and profitability only to the extent that manufacturing, marketing, and engineering contribute capabilities that make the products in the broader line more desirable to the customer than the competitions’ products. Fiegenbaum and Karnani ( 1991) analyzed data on 83 industries to study the differences between small and large firms for volume flexibility. They concluded that small firms tend to show more volume flexibility than large firms and that small firms are able to trade cost inefficiency with volume flexibility to increase their profits. Tombak ( 1988) pursued the question of whether flexibility affectsa firm’s performance by using a sample of 1,445 businessunits drawn from profit impact of market strategy (PIMS) database.He examined the relationship between manufacturing flexibility and a firm’s performance in the growth and mature phasesof the product life cycle. He found that flexibility was an important explanatory variable for a firm’s performance. Tombak and De Meyer ( 1988) acknowledged the multidimensional nature of flexibility by arguing that firms planning to introduce FMS should be concerned with both “mix flexibility” and the flexibility needed to accommodate the variance in inputs to the production process. Recently, Suarez, Cusumano, and Fine ( 1992) conducted a field study using 3 1 printedcircuit board (PGB) assembly organizations. They concluded that increase in mix and new product flexibility does not lead to higher costs or lower quality. These studies tend to view flexibility as a unidimensional concept. Jaikumar ( 1986) , for example, implicitly refers to flexibility as the ability of a system to produce a wide variety of parts. Using
STATEGY,
FLEXIBILITY,
AND
PERFORMANCE
RELATIONSHIPS
207
Browne et al.% ( 1984) taxonomy, this flexibility can be termed as “mix flexibility.” Yet this is merely one of the different types of flexibility available to a firm. Fiegenbaum and Karnani ( 1991) consider only volume flexibility. On the other hand, Suarez, Cusumano, and Fine ( 1992) did examine the multidimensional nature of flexibility. However, the special characteristics of the PCB industry and the relatively small size of the sample may prohibit us from generalizing their results to other industries. More specifically, these studies did not examine how other dimensions of flexibility are related to performance, nor did they consider business strategy, which in turn is hypothesized to directly affect performance (Swamidass and Newell 1987) and flexibility (Jaikumar 1986). 2. Background In this section we review the literature relevant to each of the constructs used in this paper. 2.1. Business Strategy Business strategy can be viewed as a part of the widely accepted hierarchy of strategy suggested several researchersincluding Hofer and Schendel ( 1978) and Fine and Hax by ( 1985). This hierarchy can be visualized to have three levels: corporate-level strategy, business-levelstrategy, and functional-level strategy. Corporate-level strategy formulation in conglomerate diversified firms is mainly characterized by the consideration of scope and resourcedeployments. At the businesslevel, the scopeand boundaries of each business unit (SBU) and the operational links with corporate strategy are specified. The basis on which the business unit will achieve and maintain a competitive advantage within its industry is also established.At the functional level [e.g., marketing strategy,manufacturing strategy, and research and development (R&D) strategy], the objectives are to support the desired business level strategy in a manner that will provide a competitive advantage and to determine how the functional level strategieswill complement each other. Recently, Venkatraman ( 1989) categorized the literature on business strategy (henceforth termed asstrategy) measurement approachesinto three types: ( 1) narrative approach, (2) classificatory approach, and ( 3) comparative approach. The narrative approach is basedon a premise that strategy of an organization is unique and should only be described in its holistic and contextual form ( Andrews 1971). The implication of this approach is that strategy can (and should) be best described verbally. Any attempt to develop a measurement schemewill be incomplete. This view of strategy measurement may have a role in the conceptual development; it has limited use for testing theories (Venkatraman 1989). The classificatory approach attempts to uncover underlying dimensions of strategy that are generalizableacrossseveralfirms. This attempt has led to an operational definition in terms of strategy typologies basedon conceptual classifications (for example, seeHofer and Schendel 1978; Miles and Snow 1978; Porter 1980) and strategy taxonomies based on empirical classifications (for example, seeHambrick 1984). The comparative approach attempts to identify and measure key traits of strategy constructs. Consequently, the focus is lesson categorization into one cell of the typology or taxonomy but on measuring the differencesalong a set of characteristicsthat collectively describe the strategy construct (for example, seeVenkatraman 1989). According to Venkatraman and Grant ( 1986) , most of the strategy construct measures are either nominal scaleswith questionable measurement properties or multi-item scales whose measurement properties have not been systematically assessed.Venkatraman ( 1989)) using the comparative approach, developed the theoretical underpinnings of the strategy construct for business strategy by proposing six dimensions of strategy: aggressiveness, analysis,defensiveness, futurity, proactiveness,and riskiness. This dimensionality
208
YASH
P. GUPTA
AND TONI
M. SOMERS
was derived and tested by treating the individual dimensions as a building block. Venkatraman ( 1989) suggestedthat his measure for strategy could be used to test theoretical relationships. In this paper, we have used the measures of strategy provided by Venkatraman ( 1989). Definitions of the six strategy dimensions are as follows: Aggressiveness: This dimension reflects the posture adopted by an organization in allocating its resourcesfor improving market positions at a relatively faster rate than the competitors in its chosen market. Analysis: This dimension refers to the tendency of an organization to search deeper for the roots of problems and to generate the best possible solutions alternatives. It also includes the extent to which an organization usesappropriate management systemssuch as information and control systemsand managerial reward systems. Defensiveness: This dimension captures the defensive behavior of an organization through the extent to which the organization employs cost reduction and efficiency seeking methods. Futurity: This dimension reflects temporal considerations embedded in key strategic decisions, in terms of relative emphasis of effectivenessconsiderations versus efficiency considerations. Emphasis on basic research, for example, is most likely to have longer term focus than application-oriented research programs that reflect shorter term focus. This aspectis operationalized by emphasizing salesforecasting and customer preferences as well as tracking of environmental trends. Proactiveness: This dimension reflects proactive behavior about participation in emerging industries, continuous search for market opportunities, and experimentation with potential responsesto changing environmental trends. Riskiness: This dimension captures the extent of riskiness in various resourceallocation decisions as well as choice of products and markets. 2.2. Manufacturing Flexibility A number of classifications of manufacturing flexibility and measuresare given in the literature (Mandelbaum 1978; Kusiak 1985; Frazelle 1986; Upton and Barash 1988; Gupta and Goyal 1989; Hutchinson and Sinha 1989; Taymaz 1989). A most thorough classification was proposed by Sethi and Sethi ( 1990) who identified 11 different flexibility dimensions. Gupta and Somers ( 1992) developed an instrument to measure manufacturing flexibility and conducted an empirical study to validate the dimensions of flexibility identified by Sethi and Sethi ( 1990). Their study revealed that 11 factors of Sethi and Sethi can be collapsed into 9 measuresof flexibility. To our knowledge, no study has identified various dimensions of flexibility. Using a field study drawn from a large sample of firms, we chose to use the measuresof flexibility suggestedby Gupta and Somers. These measuresare as follows: Machineflexibility deals with the variety of operations that the machine can perform without incurring high costs or expending a prohibitive amount of time in switching from one operation to another. Machine flexibility allows small batch sizes.This yields lower inventory costs, higher machine utilizations, the ability to produce complex parts, and improved product quality. Material-handlingflexibility is defined as the ability of material-handling systems to move different part types effectively through the manufacturing facility. This includes loading and unloading of parts, intermachine transportation, and storage of parts under various conditions of the manufacturing facility. In the end, material-handling flexibility may increase machine availability and reduce throughput times. Processflexibility is defined as the ability of a manufacturing system to produce a set of part types without major setups, something referred to as mix flexibility by Gerwin ( 1987) and Carter ( 1986). Process flexibility is useful in reducing batch sizes and, in
STATEGY,
FLEXIBILITY,
AND
PERFORMANCE
RELATIONSHIPS
209
turn, inventory costs. Becauseit allows machines to be shared, it minimizes the need for duplicate machines. Routingflexibility refers to the ability of a manufacturing system to produce a part by alternate routes through the system. The purpose of routing flexibility is to continue to produce a given set of part types, albeit at a lower rate in the event of unexpected machine breakdown. It allows for efficient scheduling of parts through improved balancing of machine loads. Volume flexibility is the ability of a manufacturing system to be operated profitably at different overall output levels, thus allowing the factory to adjust production within a wide range. Program flexibility is the ability of the system to run virtually unattended for a long enough period. Program flexibility reduces the throughput time through reducing setup times, improving inspection and gauging, and better fixtures and tools. Product andproductionflexibility is the universe of part types that the manufacturing system can produce without adding major equipment and the easewith which new parts can be added or substituted for existing parts, i.e., the easewith which the current part mix can be changed at relatively low cost in a short period. This type of flexibility is dependent on several factors: variety and versatility of available machines, flexibility of material-handling systems,and the factory information and control systems. Marketflexibility can be defined as the easewith which the manufacturing system can adapt to changing market environment. It allows the firm to respond to changeswithout seriously affecting the business and to enable the firm to outmaneuver its less flexible competitors in exploiting new business opportunities. Expansion and marketflexibility is the extent of overall effort needed to increase the capacity and capability of a manufacturing system when required. This flexibility may help shorten implementation time and reduce cost for new products, variations of existing products, or added capacity. 2.3. Organizational Performance Severalauthors have emphasized the importance of the performance concept and have provided prescriptions for improving organizational performance (for example, seeNash 1983). Yet the debate on issues of terminology, levels of analysis, and conceptual basis for assessment performance continues to rage in the academic community (Ford and of Schellenberg 1982). Some researchershave expressedfrustration with the lack of agreement on basic terminology and definition ( Kanter and Brinkerhoff 1981) . Venkatraman and Ramanujam ( 1986) suggest that business performance is a subset of the overall concept of organizational effectiveness.They argue that in its narrowest sense,business performance is associatedwith simple outcome based financial indicators (referred to as financial performance) “that are assumedto reflect the fulfillment of the economic goals of the firm.” The financial performance has been most widely used to determine organizational health of a firm. Typical indicators include return on investment, return on sales,and return on equity. A broader conceptualization of businessperformance includes emphasis on indicators of operational performance (i.e., nonfinancial) in addition to financial indicators. Under this conceptualization, measuressuch as market share, new product introduction, product quality, and market effectiveness might be considered within the domain of businessperformance. According to Venkatraman and Ramanujam ( 1986, p. 804) “the inclusion of performance indicators takes us beyond the black-box approach that seemsto characterize the exclusive use of financial indicators and focuses on those key operational successfactors that might lead to financial performance.” The sources of data to measure financial performance, operational performance, or both can be primary (e.g., data collected directly from organizations) or secondary (e.g., data from publicly available records). However, Venkatraman and Ramanujam ( 1987)
210
in
YASH
P. GUPTA
AND TONI
M. SOMERS
their study of 207 senior-level managers showed that managers tend to be less biased in their assessments their organizational performance than researchershave tended to of give them credit for. They argue that the perceptual data can be employed as acceptable operationalizations of organizational performance. Similarly, they demonstrate that a positive and statistically significant association exists between primary and secondary businessperformance data.
3. Hypotheses
In this section we justify the relationship between business strategy, manufacturing flexibility, and organizational performance. Our adaptation of Gerwin’s ( 1993) conceptual model, shown in Figure 1, makes explicit the expected links among the three variables: businessstrategy, manufacturing flexibility and organizational performance. For the purpose of simplicity, the diagram shown in Figure 1 does not show all direct and indirect paths expressedby the model. In this model, business strategy is equated with various types of manufacturing flexibility and posited as a primary influence on an organization’s performance. Specifically, the recursive causal chain shown in Figure 1 hypothesizes that business strategy will trigger the development and implementation of manufacturing flexibility dimensions. The introduction of manufacturing flexibility enhances the organization’s performance. Gerwin’s conceptual framework describes the effects of uncertainty on manufacturing strategy, flexibility, and firm performance. In this study we have used business strategy instead of manufacturing strategy. Our reasoning is based on the fact that manufacturing flexibility is not only an element of manufacturing strategy but also related to marketing and R&D strategies(Hyun and Ahn 1990; Sethi and Sethi 1990). According to Milgrom and Roberts ( 1990), it may be unprofitable for a firm to enhance manufacturing flexibility without changing its marketing strategy or to alter its marketing approach without adequate manufacturing flexibility, and yet, it may be highly profitable to do both together. One can, for example, think of several situations where a
BUSINESS STRATEGY
b
MRNUFACTUFlIN6 FLEXIBILITY
A
4
b
ORGANIZATIONflL PERFORMANCE FINANCIAL 0 GROWTH
FIGURE 1. Conceptual Framework.
STATEGY,
FLEXIBILITY,
AND
PERFORMANCE
RELATIONSHIPS
211
firm’s manufacturing function has done a superior job of developing process,expansion, and volume dimensions of flexibility. However, the marketing function did not take advantage of the resulting opportunities, because of lack of ability or, astonishingly, because marketing remained unaware of manufacturing’s capabilities (Cleveland, Schroeder, and Anderson 1989; Vickery, J%oge,and Markland 1993). This situation may result in manufacturing flexibility having a limited impact on a firm’s performance. Successful manufacturing firms exploit complementarities that exist between various functions, such as marketing, production, engineering, and organizational variables. The exploitation of complementarities among functions is usually dictated by a coherent businessstrategy of the firm (Milgrom and Roberts 1990). Moreover, several researchers have arguedthat an organization should develop a manufacturing strategythat is consistent with and linked to its business strategy (e.g., seeHayes and Wheelwright 1984). A study by Richardson, Taylor, and Gordon ( 1985) provided further evidence of the need to better understand the linkage between corporate strategy and manufacturing strategy.A field study of 64 Canadian electronics companies led the researchers conclude to that companies with a strong match between their business mission and manufacturing tasks are more profitable. They argue that the way a firm competes in its markets is a key element in determining the corporate mission. Performance will be suboptimal unless a proper congruence is achieved between the corporate mission and manufacturing strategy. Cleveland, Schroeder, and Anderson ( 1989) used a sample of six hard goods manufacturers to examine the relationship between businessstrategy, production process,production competence, and business performance. Production competence was defined as a variable rather than a fixed attribute and measuredfor how well manufacturing strengths and weaknessescomplement the priorities of the business strategy. The study showed that production competence is linked with business performance. Swamidassand Newell ( 1987), using data gathered from 35 manufacturers, studied the relationship between environmental uncertainty, manufacturing strategy,and business performance. They defined manufacturing strategy asconsisting of two types of variables: (i) content variables, such as flexibility and (ii) process variables, such as the role of manufacturing managers in strategic decision making ( RMMSDM). From this they concluded that: (a) greater flexibility leads to better performance; (b) RMMSDM is a function of environmental uncertainty and higher levels of RMMSPM result in improved performance; and (c) an organization may be better able to cope with high uncertainties by increasing manufacturing flexibility and maintaining and ensuring the RMMSDM . Schroeder, Anderson, and Cleveland ( 1986) advocated that manufacturing strategy is a processin which business strategy determines manufacturing mission and distinctive competence. By interacting between adjacent levels of the strategy hierarchy, this process determines manufacturing objectives and leads to the formulation of manufacturing policies. Recently, Gupta and Lonial ( 1994), using data from 175 manufacturing firms concluded that the linkage between manufacturing strategiesand business strategiesis a significant predictor of organizational performance. The above discussion indicates that manufacturing flexibility is one of the elements of manufacturing strategy. The congruence between manufacturing strategy and business strategy leads to increased organizational performance (Figure 1) . Therefore, it may be appropriate to examine the relationship between business strategy and manufacturing flexibility. 3.1. Manufacturing Flexibility and Business Strategy Hayes ( 1981) suggestedthat when organizations produce more products, they experience a higher level of demand variability. Similarly, the supply variability will be dependent on the number of unique parts sourced, the number of suppliers used, and the
212
YASH
P. GUPTA
AND
TONI
M. SOMERS
introduction of new or alternate materials. An organization that operates in an industry that is facing an increasingly rapid pace of technological evolution and, consequently, a shortening product life cycle and higher product turnover would be faced with additional variability. Higher levels of labor turnover, absenteeism, and low equipment reliability are additional sources of variability. To deal with these sources of variability, the organization must increase its manufacturing flexibility. When a firm decides to compete in the high end of markets, generally requiring more customized products, it would clearly need more process flexibility. Similarly, should a firm decide to compete in several related industries, it may require more processflexibility than when competing in a single industry. When competitors of an organization are constantly introducing many new and improved products embracing a wide variety of features, almost all dimensions of flexibility would be required to meet competitors’ challenge ( Krafcik 1988;MacDuffie 1991; Suarez, Cusumano, and Fine 1992). The above discussion suggeststhat: l HI : Business strategy has direct effects on the adoption of manufacturing flexibility dimensions. HI would be expected to hold whether business strategy’s effect on manufacturing flexibility results from management’s intention to pursue a competitive advantage or from a strategic necessity. The literature supports the position that business strategy incorporates the required flexibility (Slack 1987; Swamidassand Newell 1987). 3.2. Manufacturing Flexibility and Organizational Performance Kekre and Srinivasan ( 1990) empirically examined the relationship between breadth of product line (process flexibility or mix flexibility as termed by several authors) and market success.They found that broader product lines result in larger market share and profitability and that it does not seem to be associatedwith higher costs. Suarez, Cusumano, and Fine ( 1992), by examining the works of Stigler ( 1939) and Marschak and Nelson ( 1962), suggestedthat in some casesvolume flexibility may be associatedwith higher costs and/or lower quality levels. They advocated that a plant that can shrink and expand production volume and can still keep its costslow and quality high would reap higher levels of performance. Another dimension of flexibility that was studied by Suarez, Cusumano, and Fine ( 1992) is called new product flexibility (termed as product flexibility in this paper). They defined this as the ability to create new products quickly. Drawing from the works of Clark and Fujimoto ( 1991)) Suarez, Cusumano, and Fine ( 1992) argued that as technology advances rapidly and customers become more sophisticated, rapid product introduction can give firms a real competitive advantage,i.e., the organizational performance could be significantly better than its competitors. Moreover, successfulfirms that pursue different businessstrategiesmay score higher on one performance dimension or another. Firms pursuing aggressivestrategiesand higher levels of flexibility, for example, may be expected to score more highly on growth performance than on financial performance. This causal logic suggeststhe following hypotheses: l HZ: Manufacturing flexibility dimensions have direct effects on an organization’s growth ( financial ) performance. H2 provides a test of the relationship between manufacturing flexibility dimensions and an organization’s performance, and similar to previous studies (Kekre and Srinivasan 1990, Fiegenbaum and Kamani 1991), the hypothesis suggestsa direct association between the variables. l H3: Besidesdirect effects, business strategy also indirectly affects an organization’s growth (financial) performance through its effect on manufacturing flexibility dimension.
STATEGY,
FLEXIBILITY,
AND
PERFORMANCE
RELATIONSHIPS
213
H3 is formulated basedon the recent work of Gerwin ( 1993) who through his conceptual model, identified the need for an applied research agenda involving manufacturing flexibility and other related variables: Specifically, research attention needsto be devoted to studying the interrelationship among business strategy, manufacturing flexibility, and organizational performance. This hypothesis proposes that business strategy’s positive effecton an organization’s performance can be enhanced by linking it with the appropriate manufacturing flexibility dimension. From H3 it follows that (i) if the required manufacturing flexibility dimension is inappropriate to the businessstrategy,the indirect effects may be negative, even when the direct effectsof business strategy are positive and (ii) if the business strategy is inappropriate and direct effects on organizational performance are negative, then indirect effects through manufacturing flexibility could exaggerateor equalize businessstrategy’sdirect effects.To our knowledge, this hypothesis has not been examined in prior statistical research in manufacturing flexibility. In this study, path analysis provides a holistic approach for evaluating H, , Hz, and H3 in an integrative framework (Figure 2).
4. Method
In this section we provide an exposition of sample and data collection procedures and measurement of variables used in the study.
4.1. Sample and Data Collection
A total of 1,600 questionnaires were mailed nationwide to five types of U.S. manufacturing firms with 250 or more employees obtained from a mailing list directory: precision machinery, electrical and electronics, industrial machinery, metal products, and automobile and auto part firms. These types of firms were selected based on standard industrial codes (SIC’S).We chose these types of firms (industry segments) becausethey have acknowledged adoption of advanced manufacturing processes. After 2 weeksof initial mailing, a reminder letter with a fresh copy of the questionnaire was mailed to the nonrespondents. As a result of a carefully designed questionnaire, administered according to Dillman’s suggestions( 1978), the sample was believed to be characteristic of firms in their respective industries. A single respondent per organization, typically holding the title Of CEO, President, or Vice Resident of Manufacturing completed the survey instrument. The manufacturing firms were located nationwide. Of 1,600 questionnaires distributed, 279 were completed and returned, representing a responserate of 17.4%.This is a typical response rate for studies of this kind (Magal, Carr, and Watson 1988). The exclusion of questionnaires with incomplete data resulted in a final sample of 269 firms. A copy of the questionnaire is available from the authors. The rationale for selecting the five types of manufacturing firms was to create a diverse sample from which it would be easier to generalize the results. We tried to sample randomly within each industry; however, nonresponse, even with follow-ups, resulted in unequal response among the firm types. The sample is composed of large firms, with 56.4% of the firms having annual sales between 25 1 and 1,000 million dollars and 4 1.1% having 50 l- 1,000 total employees. Table 1 presents a profile of the sample. The extent of differences between respondents and nonrespondents can seldom be determined. Sometimes, however, one can make very limited checks for differences. In light of the low response rate, tests for nonresponse bias were considered necessary.To assess whether the firms included in the study were representative of those firms in the database,a comparison for selecteddemographic characteristics suggests that there were no significant differencesbetween the groups along this setof characteristics.The similarity of respondents versus nonrespondents was assessed using the Kolmorgorov-Smimov (K-S) test. None of the test statistics were statistically significant at p < 0.05.
214
YASH P. GUPTA AND TONI M. SOMERS
S,
S,
EQdoeenousVg&MgS
Ft _ExpemioalMarlret Flexibility F2- Material Handling Flexibility F3- Routing Flexibility F4- Machine Flexibility F5- Market Flexibility P6- Prcduct/Pmduction Fhibility F7- Process Flexibility Fg- Programming Flexibility F9- Volume Flexibility Pp Financial Perfmnance PO cimti Performance -
St - Aggres8ivems.s S2- Analysis S3- Defensivenes S4- Futurity s5- Pmmivencss S6- Riskiness Thesingle-headed representcausal behveen armws the tii these variables. simplicity, all linksare For not shown the in diagram, werecognized and examinedthis but toexist are in Paper.
2. FIGURE Proposed Path Analytic Model Of The BusinessStrategy, Manufacturing Flexibility, and Organizational Performance Relationship.
4.2. Operational Measures of the Variables In this section we will describe items used in measuring the variables used in our study.
4.2.1 BUSINESSSTRATEGY. A 2 l-item scale, extracted from the previous work on business strategiesby Venkatraman ( 1989), was used to measure six types of manufacturing strategies: ( 1) aggressiveness,(2) analysis, (3) defensiveness, (4) futurity, (5) proactiveness, and (6) riskiness. Respondents were asked to indicate the extent of their agreement or disagreement with each of the items using a five-point scale ranging from l-strongly agreeto 5-strongly disagree. An overall summary measure for each of the six
STATEGY, FLEXIBILITY,
AND PERFORMANCE RELATIONSHIPS TABLE 1
Sample Profile
215
Title of the respondents, % Vice President or Director of Manufacturing Plant Manager, CEO or COO Vice President or Director of Technology/Engineering Vice President of Planning Vice President or Director of Other Areas Vice President or Director of Marketing Vice President or Director of R&D Range of salesof the organizational unit, % Less than $25 million 25-50 m 51-100 m 101-250 m 251-500 m 501-1000 m Over 1000 m Total number of employees, % Under 10 1l-50 51-100 101-250 25 l-500 501-1,000 1,oo l-5,000 5,001-10,000 Over 10,000 Businesscategory, % Precision machinery Electrical and electronics Industrial machinery Metal products Automobile and auto parts Others * n = 269.
59.3 14.6 13.8 4.8 4.8 1.9 0.7 4.1 6.7 15.3 13.4 22.4 34.0 4.1 0.0 1.1 1.1 4.1 21.3 41.1 25.7 3.0 2.6 15.8 29.9 17.9 17.9 14.9 3.6
business strategieswas calculated as the average of the items in their respective strategic dimension. The scalehas been found to have high internal consistency reliability in prior empirical studies (Venkatraman 1989). The original group of items was subjected to reliability analysis, and five items were removed resulting in a 24-item scale. Appendix 1 provides the items that were removed. The six business strategies, represented as endogenous variables in the path model, their empirical indicators, and Cronbach’s coefficient alpha are portrayed in Table 2. 4.2.2. MANUFACTURING FLEXIBILITY. In a previous study, the authors developed an instrument for measuring and analyzing manufacturing flexibility (Gupta and Somers 1992): Thirty-four items affecting manufacturing flexibility were identified from the literature, and a preliminary instrument was created to measure them (see Appendix 2 for a list of the items). Respondents were asked to indicate the extent of their agreement or disagreement with each of the items on a five-point Likert-type scale ranging from l-strongly agreeto 5-strongly disagree.The final instrument, after modification, contained 21 items and had an alpha coefficient of 0.888. The large number of items required that they be condensed into a meaningful and manageabledata set. This was accomplished using principal components factor analysis. This technique assumesthat related variables describe a common underlying manufac-
216
YASH P. GUPTA AND TONI M. SOMERS TABLE 2
Business Strategy Constructs, Their Indicators, and Standardized Alpha Coeficient
Indicators We sacrifice short-term profitability to gain market share. We have cut prices to increase our market share. A strong preference for setting prices below the competition. Seeking market share positions at the expense of cash flow and nrofitabilitv. We emphasize effective coordination among functions (e.g., operations and marketing). When confronted with a major decision, we usually try to develop thorough analyses. Planning techniques (PIMS models, portfolio models). Regular manpower planning and performance appraisal. Outputs of management information and control systems. Cost control systems for monitoring performance. Emphasis on product quality through the use of systems such as quality circles. Project management techniques (e.g., PERT/CPM). Our criteria for resource allocation generally reflects shortterm considerations. Forecasting key indicators of operations. Formal tracking of significant general trends. “What-if” analvsis of critical issues. We constantly seek to indentify new opportunities closely related to our operation. We are usually the first ones to introduce new products or services in our markets. We are constantly on the lookout for business units we that can acquire. Operations in the later stages of the life cycle are strategically eliminated (i.e., liquidated, diversted). We adopt a rather conservative view when making major decisions. Approval of new projects on a “stage-by-stage” basis rather than with a “blanket approval.” Operations have generally followed the “tried-and-true” paths. A strong tendency to support projects where the expected rates of return are certain.
Constructs
Alpha Coefficients
0.66
+G+
Defensiveness
0.59
0.58
0.65
0.61
Wing flexibility dimension. Table 3 presents details of this analysis. As Table 3 shows, the 21 items yielded a 9 multi-item factor structure of manufacturing flexibility that explained 72% of the variance in the data. Each factor represents a unique flexibility dimension whose meaning is based on the items that make it up. These nine factors and Cronbach’s coefficient alphas are portrayed in Table 4. The reliability coefficients of each factor ranged from 0.70 to 0.89. Furthermore, the 21-item instrument had a criterionrelated validity of 0.7301, which correlated significantly with a separatemeasureof overall flexibility. In this study, the firm’s overall manufacturing flexibility was assessed asking by respondents to rate “To what extent is your manufacturing flexible?” This criterion question employed a five-point scale: 1 = highly inflexible, 2 = generally inflexible, 3 = neutral, 4 = generally flexible, and 5 = highly flexible. The correlation of each of the nine factors with the criterion ranged from 0.55 to 0.68.
STATEGY, FLEXIBILITY,
AND PERFORMANCE RELATIONSHIPS
217
TABLE 3 Rotated Factor Matrix of 21-Item Instrument to Measure Man@facturinx Flexibility Factor Number Item J I K L A D Z AA BB Q P DD cc E F S G X Y H G N 1 0.76114 0.12991 0.69472 0.63127 0.61855 0.53061 0.32194 0.79452 0.77289 0.57358 0.72849 0.72985 0.82217 0.7985 I 0.36621 0.83683 0.6854 0.83929 0.52378 0.86753 0.53824 0.58261 0.32588 0.8613 0.32385 0.31314 2 3 4 5 6 7 8 9
Factor Name: 1. Expansion/Market Flexibility; 2. Material Handling Rexibility; 3. Routing Flexibility; 4. Machine Flexibility; 5. Market Flexibility; 6. Product/Production Flexibility; 7. Process FIexibility; 8. Programming Flexibility; 9. Volume Flexibility. Note: Factor loadings provided by principal components analysis (varimax rotation) accounted for 72% of the total variance in the data. EigenvaIues exceeded 1.Ofor all factors except factors 8 and 9 which were 0.999.
Nunnally ( 1978) suggestedthat Cronbach’s alpha value of 0.7 is considered adequate for internal consistency reliability. For relatively new scales,however, Nunnally suggested that an alpha value of 0.6 is acceptable. Srinivasan ( 1985), on the other hand, argued that alpha value of 0.5 or higher should be considered sufficient in exploratory research. Moreover, alpha is a function of the number of items in the composite and tends to be conservative. On the basis of these observations, we decided those alpha values over 0.55 were acceptable.
4.2.3. ORGANIZATIONAL PERFORMANCE. An organization’s performance is a multifaceted construct that defies measurement by a single number. Venkatraman ( 1989) used two dimensions, growth performance and financial performance, to measure organizational performance. Each dimension of performance was measured by multiple items adapted from Gupta and Govindarajan ( 1984) (see Table 5 ). Two dimensions of performance, representedas endogenous variables, were measured by asking respondents to indicate their organization’s level of performance relative to their competition with respectto each of the six criteria. The responseoptions, anchored on a five-point Likerttype scale,ranged from ( 1) much worse than competition to ( 5) much better than competition. To assess reliability of these criteria, we calculated Cronbach’s ( 1951) coefthe ficient alpha. Two performance dimensions, suggestedby Venkatraman ( 1989), were calculated for each firm by averaging the responsesfor their respective indicators. In Table 5, we present the performance variables.
218
YASH P. GUPTA AND TONI M. SOMERS TABLE 4
Manufacturing Flexibility, Their Indicators, and Standardized Alpha CoefJicients
Item
Indicators
Constructs
Alpha Coefficients
.I. Time that may be required to double the output of the system is likely to be extremely low. I. Cost of doubling the output of the system is likely to be extremely low. K. The capacity (e.g., output per unit time) of the system can be increased when needed with ease. L. The capability (e.g., quality) of the system can be increased when needed with ease. A. Time required to introduce new products is extremely low. D. Time required to add a unit of production capacity is extremely low. Z. The ability of material-handling systems to move different part types for proper positioning and processing through the manufacturing facility is extremely high. AA. The ratio of number of paths the material-handling systems can support to the total number of paths is very high. BB. The material-handling system can link every machine to everv other machine.
Q. Decrease in throughput because of a machine breakdown is extremely low. P. Cost of the production lost as a result of expediting a preemptive order is extremely low. DD. The number of different operations that a typical machine can perform without requiring a prohibitive cost in switching from one operation to another is very high. CC. The number of different operations that a typical machine can preform without requiring a prohibitive time in switching from one operation to another is very high. E. Shortage cost of finished F. Cost of delay in meeting customer orders is extremely low. S. Number of new parts introduced per year is very high. G. Size of the universe of parts the manufacturing system is capable of producing without adding major capital equipment is extremely large. X. Changeover cost between known production tasks within the current production program is extremely low. Y. The ratio of the waiting costs of processed parts and the total output is extremely low. H. The manufacturing system is capable of running virtually unattended during the second and third shift.
Routing flexibility
Machine flexibility
0.85
0.81
0.89
0.85
0.70
profitably is extremely high.
Volume flexibility
STATEGY,
FLEXIBILITY,
AND
PERFORMANCE
RELATIONSHIPS
219
TABLE 5 Financial and Growth Performance Indicators and Standardized Alpha Coefficient Alpha Coefficients
Indicators
Constructs
5. Model Results In this study, path analysis is the technique we use to explore our propositions about the sequential relationship among business strategy, flexibility, and organizational performance with complex direct and indirect effects,for example, such as those in H3. The most direct conversion of the conceptual model (see Figure 1) involved the specification of a path model (see Figure 2) that would allow for the examination of both the direct and indirect effects of strategy and flexibility on organizational performance. It must be pointed out that Figure 2 is simplified to show only a few of the direct and indirect strategy-organizational performance connections, and it does not show all the paths tested in accord with the effects anticipated in our hypotheses. The model is recursive insofar as it is assumedthat reciprocal causation in the form of causal feedback loops does not exist (Kerlinger and Pedhazur 1973). While performance may affect future business strategy and/or flexibility adoption, these effects are best modeled by considering time lags and therefore are outside the scope of this research. Although this assumption may not be totally justified, casual feedback loops have not yet been demonstrated empirically in this literature and would require complex estimation techniques (Cohen and Cohen 1983). In this model, our hypotheses are made more explicit through the construction of an arrow diagram depicting the expected causal sequence. The resulting diagram shows businessstrategyas an exogenousvariable affecting the required manufacturing flexibility and organization’s performance. Manufacturing flexibility and organizational performance are endogenous variables that are explained partly by business strategy, and in the caseof performance, manufacturing flexibility. Variables are classified as exogenous (independent) to the model when they only emit arrows; endogenous (dependent) variables receive arrows. Each arrow is called a path, and paths connecting the variables can be either direct or indirect. Both direct and indirect strategy and performance connections are shown in the diagram, and the paths are drawn in accordance with the effectsanticipated in the hypotheses.Direct paths are those connections between variables comprised of a single pathway. A standardized path coefficient, or beta, indicates the direction (either plus or minus) and magnitude of influence between variables. For example, the standardized path coefficient between manufacturing flexibility and organization’s performance is the standardized beta weight resulting from the prediction of performance from manufacturing flexibility with business strategy already partialled out.
220
YASH
P. GUPTA
AND
TONI
M. SOMERS
Indirect paths are compound pathways (with mediating variables) made up of several direct pathways. We expect there to be an indirect path from business strategy to performance that is comprised of several significant direct paths. To test Hi, the regression results and the standardized path coefficients (pc.) representing the direct effects of business strategy on manufacturing flexibility are shown in Table 6. The effects of business strategy on manufacturing flexibility are indeed quite distinguishing. Two strategies,aggressiveness proactivenesswere significant predictors and of several types of flexibility. On the other hand, the strategies of a business oriented toward futurity or riskiness were not found to have any significant impact on type of manufacturing flexibility. An aggressivestrategy was a significant determinant of all types of manufacturing flexibility. A proactive strategy emerged as a direct significant predictor of expansion and market, material handling, routing, machine, market, programming, and volume flexibility. Four strategies had significant direct effects on expansion and market flexibility: aggressiveness(p.c. = 0.313, p < 0.01); analysis (p.c. = 0.139, p c 0.10); defensiveness(p.c. = -0.149, p < 0.05); and proactiveness (p.c. = 0.133, p -C 0.05). In fact, a defensive strategy was significantly related to only this type of flexibility (p.c. = -0.149, p < 0.05). The negative coefficient suggests with increased that emphasis on cost reduction and efficiency-seeking methods, the lesslikely the firm’s need to increase the capacity and capability of its manufacturing system. An analysis strategy was significantly associated with both expansion and market (p.c. = 0.139, p < 0.10) and product and production flexibility (p.c. = 0.149, p < 0.05). This suggeststhat an increase in a firm’s overall problem-solving posture is accompanied by increases in (i) capacity and capability of its manufacturing system when needed and (ii) part types that the system can produce without requiring additional major equipment. The models differed in their ability to explain variance in type of manufacturing flexibility. An examination of the adjusted R2 statistics in Table 6 indicates that a modest amount of variance in type of flexibility has been explained by businessstrategy. Business strategy as a whole explained from 1 to 17% of the variance in type of manufacturing flexibility. All models, except one, significantly demonstrated that business strategy has an impact on manufacturing flexibility. It would appear, however, that business strategy has ,no direct effect on product and production flexibility (F = 1.38, p = 0.2 195) even though aggressiveness (p.c. = 0.110, p < 0.10) and analysis (p.c. = 0.149, p < 0.10) strategies approached significance at the 0.05 level. The best models were found for
TABLE
6
Direct Effect of Business Strategy on Manufacturing Flexibility
FI Independent Variables Aggressiveness S IAnalysis S2 Defensiveness S3 Futurity S4 Proactiveness S5 Riskiness S6 Adjusted R* F-Ratio (Probability) Expansion and Market 0.3 13’ 0.139*** -0.149;. 0.066 0.133.’ 0.026 0.17 10.16. (0.~) F2 Material Handling 0.13989 0.02 -0.028 0.05 I 0.191. -0.013 0.06 3.75+ (0.0013) F3 F4 Dependent Variable9 F5 F6 Product and Production 0. I lo*** 0.149*** 0.026 -0.055 -0.0 I 5 0.036 0.0 I 1.38 (0.2 195) F-7 F8 F9
Routing 0.165’ 0.127 -0.103 0.058 0.150** -0.006 0.08 4.64; (0.0002)
Machine 0.209; 0.087 -0.083 0.065 0.190* -0.016 0.1 I 6.45* (0.oooo)
Market 0.158’ 0.025 0.004 0.02 0.159. 0.05 I 0.06 3.61* (0.0019)
Process 0.136”’ 0.012 -0.069 0.106 0.099 0.04 I 0.03 2.24” (0.0397)
Programming 0.229. 0.082 -0.064 0.017 0.158’ 0.022 0. I 5.92’ (0.~)
Volume 0.136*’ 0.002 -0.022 0.058 0.236* -0.014 0.08 4.57* (0.0002)
Note:N=269;*p<0.01;**p<0.05;***p<0.10. * Values are standardized regression coefficients.
STATEGY,
FLEXIBILITY,
AND
PERFORMANCE
RELATIONSHIPS
221
predicting expansion and market and machine flexibility. In these models, business strategy accounted for 17 and 1I%, respectively, of the variance. In a path model, the path coefficients not only identify the direct effect of each of the exogenous variables on the appropriate dependent variables, but they can also be used to calculate both the indirect and the total effects of each variable on the respective dependent variables. As seen in Table 7, the total effect is simply the sum of the direct effectsand all the indirect effectsthat occur through intervening variables. The indirect effect of a variable is that which is traceable through its association with other variables. Estimates for indirect effectsare computed by tracing in the path diagram (see Figure 2) from businessstrategy (S l-S6 ) through the nine types of manufacturing flexibility (FlF9) to performance (growth and financial). For example, the indirect effects of aggressivenessstrategy (S 1) are computed according to the “tracing rule” as: 0.03588 = 0.113(0.313) + 0.056(0.139) + 0.065(0.165) + 0.035(0.209) + 0.003(0.158) + -0.011(0.110) + -0.111(0.136) + -0.061(0.229) + 0.033(0.136) It has been suggestedthat having calculated the path coefficients for a just-identified model, path coefficients that do not meet criteria of statistical significance be deleted from the model (referred to as theory-trimming approach). Following the theory one would delete path coefficients that are not statistically significant at a prespecified level of significance. When it is desired to test whether more than one path coefficient within a given equation may be deleted, it is appropriate to test them simultaneously by using the F-test. This study, however, did not apply the theory trimming approach that would reduce the total number of variables to the smallest core of variables that could efficiently predict growth and financial performance. To test H2 and H3, the effects of business strategy and manufacturing flexibility on growth and financial performance were decomposed into direct, indirect, and spurious. Many (5 of 6) of the business strategies were found to have an impact on the firm’s financial performance. Aggressiveness(p.c. = 0.267, p < 0.01 ), analysis (p.c. = 0.183, p < 0.05), defensiveness (p.c. = -0.129, p < O.lO), proactiveness (p.c. = 0.191, p < .O.O and riskiness ( p.c. = -0.2 19,p -C0.0 1) have significant direct effectson financial 1) performance. The absolute value of the standardized path coefficient allows us to determine which strategiesare more important for financial performance. Thus, aggressiveness and riskiness strategiesare the most important predictors of financial performance. The positive coefficientsimply that financial performance increasesasa firm’s strategicposture increases.In contrast, a negative coefficient, for example, that found for riskiness strategy, suggests that financial performance decreasesas the extent of risk taking increases. Only one type of flexibility, processflexibility, significantly affectsfinancial performance (p.c. = -0.111, p c 0. lo), and its effect is modest, at best. Nonetheless, an increase in the capability of a manufacturing system to produce a set of part types without major setup is accompanied by a reduction in financial performance. The overall adjusted R2 for the model is 0.283 (p < 0.01). Surprisingly, when growth performance is considered as the dependent variable, several types of manufacturing flexibility significantly affect it. In contrast, we found that only one businessstrategy, proactiveness, appeared to be significantly related to firms’ growth performance. The growth performance of a firm is influenced by expansion and market (p.c. = 0.242, p < 0.01 ), routing (p.c. = 0.143, p -C 0.05), product and production (p.c. = -0.128, p c 0.05 ), process ( p.c. = -0.098, p < 0.10) and volume flexibility (p.c. = 0.162, p -C 0.01) as well as by a proactive business strategy (p.c. = 0.253, p -C0.0 1). Overall, almost 20% of the variance in growth performance is explained by the model (Figure 3 ) .
TABLE
7
Direct, Indirect, and Total E#ect of Business Strategy and Manufacturing Flexibility on Organizational Performance
Dependent Variable: Growth Performance
Dependent Variable: Financial Performance Total 0.302* 0.203* -0.146** -0.008 0.220; -0.224* 0.335* 0.178* -0.124** -0.010 0.218* 0.243* 0.101*** 0.032 -0.050 0.033 0.095*** -0.003 0.191* 0.059 -0.123** 0.098*** 0.348* -0.048 0.242* 0.166* 0.242* 0.090 0.027 -0.073 0.065 0.253* -0.045 0.113*** 0.053 0.033 -0.025 0.022 -0.002 -0.002 0.019 spurious
Independent Variable
Direct
Indirect
r
Indirect 0.056 0.05 I 0.014 -0.050 -0.007 0.028 Total
Direct
spurious
r
0.247* 0.1 lo*** -0.109*** 0.048 0.341* -0.020
Sl s2 s3 s4 S5 S6
0.267: 0.183+* -0.129*** -0.014 0.191* -0.219*
0.035 0.020 -0.017 0.006 0.029 -0.005
Fl
0.113
-
-
0.110*** 0.014 0.143** 0.036 0.069 0.010 0.107*** 0.046 0.028
0.352: 0.024 0.250* 0.082 0.097***
F2 F3 F4 F5 0.056 0.065 0.035 0.003 0.014 0.143** 0.036 0.069 0.066 0.197* 0.069 0.058
0.056 0.065 0.035 0.003
0.122** 0.262* 0.104*** 0.06 1
-
F6 F7
-0.011 -0.111*** -0.011 -0.111*** -0.042 -0.131** -0.053 -0.242:
-
-0.128** -0.098’::
-0.128** -0.098***
0.007 0.044
-0.121** -0.054
-0.058 0.162* -
F8 F9
-0.06 1 0.033 -0.06 1 0.033 -0.066 0.068
0.283 8.062 (0.000) -0.127** 0.101***
-0.058 0.162*
-0.047 0.100***
-0.105*** 0.262* 0.191 5.236 (0.0000)
Aggressiveness Analysis Defensiveness Futurity Proactiveness Riskiness Expansion and market flexibility Material-handling flexibility Routing flexibility Machine flexibility Market flexibility Product and Production flexibility Process flexibility Programming flexibility Volume flexibility Adjusted RZ F-Ratio (Probability) Ratio of Correlations duplicated within 0.10
13115
12115
Note: n = 269, * p -c 0.01; ** p < 0.05; *** p i 0.10.
STATEGY,
Direct
FLEXIBILITY,
AND
PERFORMANCE
RELATIONSHIPS
223
Effects of BusinessStrategy on Manufacturing Flexibility and Direct Effects of BusinessStrategy and Manufacturing Flexibility on Financial And Growth Petionnance:
sl,
s2,
s3,
.%,S6
.-b
F4
Sl, ss
F5 F8
F9
Sl,
s2
b b
Fa
Sl
Fl
F7, F9
SS
-b
Fl, F3, Fa,
BusinessStrategy Indirect Effects on Growth Performance‘Ihmugb Manufacturing Flexibility :
s1-----b
ss /
FI .-b
PC
Endo=nous SI - Expansion/Market Flexibility
Sa - Materiel Handling Ss - Routing Flexibility SI _ Machine Flexibility Ss - Market Flexibiity Ss - Product/Production Flexibility
VarMIss
Flexibility
FI - Expansion/Market Flexibility Fz - Material Handling Flexibility R - Routing Flexibility F4 - Machine Flexibility Fs - Market Flexibility F6 - Pmduct/Production Flexibility FT - Process Flexibility F8 - Programming Flexibility Fs _ Volume Flexibility PP. Financial Performance PO - Growth Perfcmnance
FIGURE 3. Summary of Significant Paths from Figure 2 Based on Standardized Regression Coefficients.
It was predicted that any relations between business strategy and organizational performance would be mediated by type of manufacturing flexibility. We expect there to be an indirect path from business strategy to growth and financial performance that is comprised of two significant direct paths. The strength of an indirect path is determined by computing the product of the standardized (direct) path coefficients that make up such a compound path. The results in Table 7 demonstrate partial support for H3, which had posited that manufacturing flexibility would play a mediating role between businessstrategy and the organizational performance of firms. The role of manufacturing flexibility asan intervening variable is confirmed by the finding that aggressive proactive business and strategies marginally influence growth performance indirectly through their effects on manufacturing flexibility. Contrary to expectations, none of the business strategies had indirect effects on financial performance through manufacturing flexibility. Business
224
YASH
P. GUPTA
AND
TONI
M. SOMERS
strategy indirect effects on financial through flexibility are quite small. These business strategiesapparently have no bearing on manufacturing flexibility. It would appear then that part of the relationship between businessstrategyand financial performance is indirect through other variables not included in this analysis. Billings and Wroten ( 1978) suggested once path coefficients have been determined, that they should be verified by attempting to recompute the correlation matrix through the calculation of the total effectsbetween the setsof related variables. In Table 7, a comparison of the estimated correlations, found by summing the direct and indirect effects (total effect), with the original correlations ( r) between the independent variables and dependent measuresprovides supporting evidence of the “goodness of fit” of these models. If we used the criterion that the absolute difference between the reproduced and original correlations should not exceed 0.10, our models can duplicate almost all ( 13 of 15) of the original correlations for financial performance and 12 of the 15 correlations involving growth performance (Namboodiri, Carter, and Blalock 1975). Further, the extent of the spurious effectsfound for both models suggests that unless business strategy is considered in these models, the direct effects of manufacturing flexibility on growth and financial performance will be underestimated. For some variables, the spurious effects and their direct effects are of opposite sign, implying that the zero-order correlation coefficient alone would provide a misleading relationship between manufacturing flexibility and firms’ performance (Table 8). A summary of the results reported in this section is given in Figure 3. 6. Discussion The objective of this paper was to determine the relationship between businessstrategy, manufacturing flexibility, and organizational performance. The data used to test our hypotheses and the development of flexibility measuresreported by Gupta and Somers ( 1992) are drawn from the same database.Therefore, findings of this research put them in more of an exploratory rather than confirmatory category. The aggressiveness dimension of business strategy is significantly related to all of the dimensions of manufacturing flexibility identified in this paper (Table 6 ) . Since aggressive organizations tend to sacrificeprofitability and set prices below competition to gain market share even at the expense of cash flow profitability and since manufacturing flexibility allows an organization to increase market share by improving its ability to produce a greater mix of products, change the volume of production, and respond to changing market conditions, it is obvious that the aggressiveness strategy should be significantly
TABLE
Zero-Order FI Expansion and Market FI F2 F3 F4 F5 F6 Fl F8 F9 F2 Material Handling Correlation F3 Matrix F4
8
Flexibility F6 Product and Production n Variables F8 F9
for Manufacturing F5
Routing
Machine
Market
Process
Programming
Volume
0.3697’ 0.2226’ 0.2131* 0.2608* 0.3893* 0.0489 0.2787* 0.3816* 0.1248" 0.1059”’ 0.3579* 0. I506** 0.2041* 0.3524* 0.2505; 0.19541 0.2816* 0.2827* 0.0084 0.3869* 0.2768* 0.1508** 0.2683* 0.2504; 0.3999* 0.3052*
0.2864 0.0637 0.2208’ 0. I954f
0.0287 0.0726 0.2501*
0.0537 0.3747s
0.2933; -
STATEGY,
FLEXIBILITY,
AND
PERFORMANCE
RELATIONSHIPS
225
related to manufacturing flexibility. These results are further supported by the findings reported in Table 7. In this table, the direct effect of aggressivelypursuing market share (aggressiveness)on growth performance is not significant; however, aggressiveness orientation affectsgrowth indirectly through manufacturing flexibility, suggestingthat firms opting for aggressiveness strategy and aiming for an increasing growth trend must enhance manufacturing flexibility. Proactive organizations generally seek new opportunities for business that can be acquired and are generally the first to introduce new products. Organizations pursuing this strategy seemto seekall the flexibility dimensions except processflexibility and product and production flexibility (see Table 6 ) . In addition, the proactiveness strategy also influences growth performance indirectly through manufacturing flexibility. One possible explanation, albeit weak, for the lack of a direct relationship between proactiveness and process and product and production flexibility dimensions is that the newly acquired businessesand their associated products are generally kept independent of the current business units. Thus, the current manufacturing plant may not have a great need for processflexibility. The path analysis results demonstrate that aggressiveness proactiveness strategies and influence growth performance indirectly through manufacturing flexibility (Table 7 ) . These findings suggestthat the linkage between strategy and growth performance is more complex than suggestedby previous researchand thus emphasize the need for additional studies rising multivariate research designs. It is interesting to observe that organizations pursuing a defensivenessstrategy tend to seekvery little manufacturing flexibility. This strategy tends to be reactive and emphasize the implementation of cost reduction and efficiency improvement methods. These characteristics generally do not require high levels of manufacturing flexibility (Giffi, Roth, and Seal 1990). The lack of a significant relationship between analysis and futurity dimensions of business strategy and dimensions of manufacturing flexibility may be attributed to the fact that defensivenessorientation in businessstrategy is strongly related to futurity and analysis (see Venkatraman 1989). Our results show that the aggressiveness,analysis, defensiveness, proactiveness, and riskiness dimensions of business strategy are significantly related to the financial performance of an organization (Table 7). The directions (signs) of all these relationships, except aggressivenessand defensiveness, are consistent with those obtained by Venkatraman ( 1989). Our results on the association between aggressivenessand financial performance, however, are in consonance with Hambrick, Macmillan, and Day’s ( 1982) findings. They suggestedthat market share gains can be achieved without loss of current profitability. One plausible explanation could be that our study deals with the data collected exclusively from manufacturing companies, whereas Venkatraman collected data from a number of industries. In any event, the inconsistencies in results suggest the need for additional field studies in this area. The significant and positive relationship between proactiveness and growth performance is supported by previous research (Venkatraman 1989). Our findings indicate that five dimensions of flexibility out of nine are significantly related to growth performance, whereas only one dimension significantly affectsfinancial performance (Table 7). It is conceivable that since manufacturing flexibility is still a relatively new phenomenon, management may not yet have recognized this linkage between flexibility and financial performance. Moreover, Kekre and Srinivasan ( 1990) suggested that growth performance (market share) should eventually stimulate financial performance of an organization. Expansion and market flexibility have a positive and significant impact on growth performance. This flexibility allows a firm to respond to changes in customers’ tastes, declining product life cycles, and uncertainty in sources of supplies and, thus, enables it
226
YASH
P. GUPTA
AND TONI
M. SOMERS
to become a time-based competitor. These findings are consistent with the literature, i.e., time-based organizations tend to perform significantly better than their competitors who do not pursue time-based strategies (Stalk and Hout 1990). To enhance this flexibility the processof production planning and inventory controls must be integrated with marketing functions, such as product development and market forecasts.Similarly, improved relationships with suppliers and well-developed distribution channels are essential (Sethi and Sethi 1990). The impact of routing flexibility on growth performance suggeststhat organizations in their attempt to enhance this flexibility must employ multipurpose machines, pool identical machines into machine groups (Stecke and Kim 1989)) improve versatility of material handling systemsand system control software (Yao 1985), plan for underutilization of machines, or build redundancy in machines so that the production system can be rescheduled and maintain the overall production rate in caseof a machine breakdown. This also suggests training programs designedto increasemaintenance workers’ that knowledge of the production system to prevent damage and to reroute production may be beneficial (Gerwin 1989). It is interesting to observe that process flexibility impacts negatively on both growth and financial performance. One plausible explanation for this result is that a large number of organizations have implemented automated manufacturing technology with the fuzzy notion to increase process flexibility. However, they have continued to use these technologies under massproduction regimes. Thus, they do not necessarily benefit from the potential increasein processflexibility while incurring the cost of investment in automated technologies (see for example, Bessant 1985; Jaikumar 1986; Majchrzak 1988; Primrose 1988). Another explanation for the negative relationship between processflexibility and performance hasbeen put forward by Gerwin ( 1993) who suggested increasedproduct that variety (process flexibility) leads to complexity and confusion that raisesoverhead costs. Contrary to expectations, product and production flexibility has a negative relationship with growth performance. As noted by Ettlie and Penner-Hahn ( 1994), the true value of this type of flexibility is not realized until the next generation of products is introduced. It is possible that a significant number of our respondents may not have launched a new generation of products since improving this dimension of flexibility. The finding that volume flexibility has a significant and positive relationship with growth performance suggeststhat workers must possess skills that can be used elsewhere when production volume decreases. This, in turn, implies that management must provide cross-functional training to their employees(Gerwin 1989). In addition, volume flexibility can be enhanced by realizing the importance of subcontracting networks and implementing just-in-time concepts. 6.1. Implications The management of manufacturing flexibility has been cited as a critical issue for both manufacturing executives and general managers (Upton 1994). The increasing beliefs in the strategic role of manufacturing in general (Hayes and Wheelwright 1984), along with the level of executive manager participation in our study, suggestthat manufacturing flexibihty will continue to be a critical issue well into the 1990s. This study was driven by a strong desire to provide empirical evidence that could form the basis for the guidelines for managing flexibility. In this article, evidence is provided that manufacturing managers should not increase all dimensions of flexibility in their drive to make improvement in their organization’s performance. Some dimensions of flexibility may have a counterproductive effect,and others may not significantly contribute to their organizational performance goals. Consequently, the investments in flexibilityenhancing mechanisms must be in consonance with the dimension of flexibility that management may be attempting to improve.
STATEGY,
FLEXIBILITY,
AND
PERFORMANCE
RELATIONSHIPS
227
Several dimensions of flexibility (for example, expansion and market and volume) allow an organization to respond to markets expeditiously. Those organizations who know their markets better may not need as much flexibility. On the other hand, less competent organizations who are unable to predict the market changes may favor flexibility. According to Ettlie and Penner-Hahn ( 1994) “these sameorganizations may have difficulty implementing flexibility and other manufacturing innovations.” Improved expansion and market flexibility provide greater freedom to experiment with variations in product designs. This, in turn, will result in escalation in transactions between manufacturing and product design functions and “may lead to greater conflict unless the additional experimentation is prevented from interfering with regular production (Gerwin 1992, p. 2 15) .” Senior executives of an organization must improve working relationships between thesefunctions by implementing mechanismsthat enhance communications and encourage input from manufacturing in product design process (Gupta and Somers 1993) . Gerwin ( 1992) suggested increasedexpansion and market that flexibility may reduce pressure on product engineers to design the products “right” the first time because the inadequacies or defects in product design may be screened and modified when prototypes are being produced or even after, causing escalation in uncertainty for manufacturing function. The above discussion of results suggeststhat the most effective way to increase manufacturing flexibility is to invest heavily in flexible manufacturing capabilities, including technology, organizational systems, and worker training (worker skills are especially important). According to Fisher, Jain, and MacDuffie ( 1994)) the companies that fail in their efforts to increase flexibility do so not because they haven’t bought the right hardware, but because they either don’t understand the importance of worker training or are hampered in their efforts to institute it.
7. Limitations and Future Research
Although this study provides interesting insights into the pattern of relationships among various dimensions of manufacturing flexibility, business strategy, and organizational performance, the results must be interpreted cautiously. It should be pointed out that although the adjusted R2 statistics of 0.283 and 0.191 in Table 7 would be considered acceptable, it must also be concluded that there are other determinants of successthat were not included in our model. These factors may include environmental uncertainty as measuredby Swamidassand Newell ( 1987) or components of manufacturing strategy other than manufacturing flexibility as recognized by Miller and Roth ( 1994) and Gupta and Lonial ( 1994). It would be worthwhile to expand our model in the future studies to include these factors. Similarly, while our results support Hi, in that the adjusted R2 statistics are significant (Table 6)) not all the coefficients in each equation turned out significant. Further, all but three of the adjusted R2 statistics are <O.lO. These results indicate that there are probably a number of other factors affecting the choice of each type of flexibility. More insight on flexibility must come from subsequent studies. Three of the six constructs of business strategy have Cronbach alpha values below 0.6, which Nunnally identified as the boundary for reliability, and three are barely above that. Although the precedents exist in the literature where alpha value of 0.5 and above wasconsideredacceptable,the low value may be attributed to the mixing of manufacturing industries in our sample. A future study may focus on a single manufacturing industry so that the above issue may be resolved. In addition, the single-manufacturing industry study would allow researchersto control confounding market variables that may vary from industry to industry. The study used a single respondent per organization. Seeking responsesfrom a single informant to make judgments on complex organizational characteristics may increase the subjective propensity of single respondents to seekout consistency in their responses
228
YASH P. GUPTA AND TONI M. SOMERS
and increase random measurement error. According to Miller and Roth ( 1994) the random error components may result from the reporting process,knowledge deficiencies, inadequate measures,or some combination of these factors. Campbell ( 1955) identified two criteria for collecting information about a social system: ( 1) the respondents should occupy roles that make them knowledgeable about the issues being researchedand (2 ) they should be willing and able to communicate with the researchers.In our study, the monorespondent problem may have been moderated by the fact that high-ranking respondents tend to be more reliable sourcesof information than their lower ranking counterparts (Phillips 1981) . Moreover, we believe that such a strategy enabled us to achieve a greater and more diverse sample size. Future studies, however, should attempt to avoid this methodological pitfall by pragmatically obtaining multiple sources of information within single organizations. This study has established the relationship between business strategy, manufacturing flexibility, and performance; it would be an important research issue to examine the impact of dimensions of flexibility on quality and productivity. Also, it would be useful to examine how various types of organization structures may have a moderating effect on the need for manufacturing flexibility. 8. Conclusions In this study we have examined the relationship between business strategy, manufacturing flexibility and performance. Our results indicate that businessstrategy contributes both directly and indirectly to organizational performance. The findings provide evidence of direct effectsof(i) businessstrategyon manufacturing flexibility, and (ii) manufacturing flexibility on organizational performance.’
’ We acknowledgethe constructive comments made by two reviewers and the associateeditor on the previous version of this paper.
Appendix 1. Business Strategy Scales
Item Our information systemsprovide support for day-to-day decision making. We have made significant modifications to our manufacturing technologies. We emphasize basic researchto provide us with future competitive edge. Our competitors generally preempt us by expanding capacity ahead of us. Our mode of operations is riskier than our competitor’s operations.
Strategy Scale Analysis Defensiveness Futurity Proactiveness Riskiness
The five items above were removed based on an examination of the item-to-total correlation data, which suggested that these items could be removed from their respective scales.Low correlations for these items suggestedthat they may not share the core of the construct.
Appendix 2. Initial List of 34 Items Measuring Manufacturing
Flexibility
and Their Source
Code A B C
Item Description Time required to introduce new products is extremely low. Cost required to introduce new products is extremely high. Time required to increase or decreaseproduction volume by 20% is extremely low.
Source Sethi and Sethi (1990) Sethi and Sethi ( 1990) Sethi and Sethi (1990)
STATEGY, FLEXIBILITY,
AND PERFORMANCE RELATIONSHIPS
229
Appendix 2. (continued) Item Description Time required to add a unit of production capacity is extremely low. Shortage cost of finished products is extremely low. Cost of delay in meeting customer orders is extremely low. Size of the universe of parts the manufacturing system is capable of producing without adding major capital equipment is extremely large. The manufacturing system is capable of running virtually unattended during the second and third shift. Cost of doubling the output of the system is likely to be extremely low. Time that may be required to double the output of the system is likely to be extremely low. The capacity (e.g., output per unit time) of the system can be increased when needed with ease. The capability (e.g., quality) of the system can be increased when needed with ease. The per unit manufacturing cost is extremely stable over widely varying levels of total production volume. The range of volumes in which the firm can run profitably is extremely low. Average number of possible ways in which a part type can be processedin the system is extremely high. Cost of the production lost as a result of expediting a preemptive order is extremely low. Decreasein throughput becauseof a machine breakdown is extremely low. Time required to switch from one part mix to another is extremely low. Number of new parts introduced per year is very high. Cost required to switch one part mix to another is extremely low. Total incremental value of new products that can be fabricated within the system for a 20% additional cost in new fixtures, tools, and part programs is extremely low. Volume (number of different part types or range of sizesand shapes)of the set of part types that the system can produce without major setups is extremely low. Extent to which product mix can be changed while maintaining efficient production is very narrow. Changeover cost between known production tasks within the current production program is extremely low. The ratio of the total output and the waiting cost of parts processed is extremely low. The ability of material handling system to move different part types for proper positioningand processing through the manufacturing facility is extremely high. The ratio of the number of paths the material handling systemscan support to the total number of paths is very high. The material handling system can link every machine to every other machine. The number of different operations that a typical machine can perform without requiring a prohibitive time in switching from one operation to another is very high. The number of different operations that a typical machine can perform without requiring a prohibitive cost in switching from one operation to another is very high. Source Sethi and Sethi (1990) Abadie et al. (1988) Abadie et al. (1988) Chatterjee et al. (1984) Jaikumar (1986) Carter ( 1986) Carter (1986) Sethi and Sethi (1990) Sethi and Sethi (1990) Falkner (1986) Sethi and Sethi (1990) Chatterjee et al. (1984) Ball ( 1989) Browne et al. (1984); Buzacott ( 1982) Browne et al. ( 1984); Buzacott ( 1982) Jaikumar (1986) Browne et al. (1984); Buzacott ( 1982) Jaikumar (1986) Gerwin (1987) Carter ( 1986) Warnecke and Stienhipler (1982) Son and Park ( 1987) Sethi and Sethi (1990) Sethi and Sethi (1990) Chatterjee et al. (1984) Sethi and Sethi (1990) Sethi and Sethi (1990)
Code
D E F G H. 1 J K L M N 0 P
Q
R S T U V W X Y Z AA BB cc DD
230
YASH P. GUPTA AND TONI M. SOMERS Appendix 2. (continued)
Code EE FF GG HH
Item Description The ratio of the total output and the idle cost of a typical machine for a given period is very low. The number of tools or the number of programs that a typical machine can use is very low. The extent of variations in key dimensional and metallurgical properties of the raw input stock a typical machine can handle is very low. The rate at which a typical machine becomes obsolete when a new product is introduced is very high.
Source Son and Park (1987) Tarondeau ( 1982) Gerwin (1987) Lam (1987)
References
ADLER, P. S. ( 1988), “Managing Flexible Automation,” California Management Review, 30, 2, 34-56. ANDREW&K. R. ( 197 1), The Concept ofcorporate Strategy, Dow Jones-Irwin, New York. BESSANT, ( 1985), “The Integration Barrier: Problems in the Implementation of Advanced Manufacturing J. Technology,” Robotica, 3, 2, 97-103. BILLINGS,R. S., AND S. P. WROTEN( 1978), “Use of Path Analysis in Industrial/Organizational Psychology: Criticisms and Suggestions,” Journal of Applied Psychology, ,63, 6, 677-688. BRILL,P. AND M. MANDELBAUM 1989), “On Measuresof Flexibility in Manufacturing Systems,”International (
Journal qfProduction Research, 27, 5, 747-756.
BROWNE, D. DUBOIS,K. RATHMILL, S. P. SETHI, AND K. E. STECKE(1984), “Classification of Flexible J., Manufacturing Systems,” The FMS Magazine, 2, I, I 14-117. BUZACOTT, A. ( 1982), “The Fundamental Principles of Flexibility in Manufacturing Systems,” Proceedings J. of the 1st International Conference on Flexible Manufacturing Systems, Brighton, UK, (October 2022).
CAMPBELL, T. ( 1955), “The Informant in Quantitative Research,” American Journal of Sociology, 60, 4, D.
339-342.
CARTER,M. F. ( 1986), “Designing Flexibility into Automated Manufacturing Systems,” Proceedings of the
2nd ORSA/TIMS Conference on Flexible Manufacturing Systems: Operations Research Models and Applications, K. E. Steckeand R. Suri (eds.), Elsevier SciencePublishers, Amsterdam, The Netherlands.
CHATTERJEE, M. COHEN,W. MAXWELL,AND L. MILLER ( 1984), “Manufacturing Flexibility: Models and A., Measurements,” Proceedings of the 1st ORSA/TIMS Conference on FMS, Ann Arbor, Ml, 49-64. CHUNG,C. H. AND I. J. CHEN( 1990), “Managing Flexibility of Flexible Manufacturing Systemsfor Competitive Edge” in The Selection and Evaluation ofAdvanced Manufacturing Technologies, M. J. Liberatore (ed.), Springer-Verlag, Berlin, 280-305. CLARK,K. B. AND T. FUJIMOTO 199 I). Product Development Performance: Strategy, Organization, and Man( agement in the World Auto Industry, Harvard BusinessSchool Press,Boston, MA. CLEVELAND, R. G. SCHROEDER, J. C. ANDERSON 1989), “A Theory of Production Competence,” G., AND (
Decision Sciences, 20, 4, 655-668.
COHEN,J. AND P. COHEN( 1983), Applied Multiple Regression/Correlation Analysis for the Behavioral Sciences, Erlbaum, Hillsdale, N.J. COX,T. ( 1979))“Toward the Measurementof Manufacturing Flexibility,” Production and Inventory Management
Journal, 30, 1, 68-72.
CRONBACH, J. ( 1951), “Coefficient Alpha and the Internal Consistency of Tests,” Psychometrika, L.
297-334.
16, 2,
CUSUMANO, ( 1988), “Shifting Economies: Craft Production to the Flexible Factory,” Working Paper 2012M. 88, Sloan School of Management, MIT, Cambridge, MA. DILLMAN, D. ( 1978), Mail and Telephone Surveys: The Total Design Method, John Wiley & Sons,New York. DIXON, J. R. ( 1992), “Measuring Manufacturing Flexibility: An Empirical Investigation,” European Journal of Operational Research, 60, 2, 131-143. ETTLIE,J. E. AND J. D. PENNER-HAHN 1994), “Flexibility Ratios and Manufacturing Strategy,” Management ( Science, 40, 1I, 1444-1454. FERDOWS, AND A. DE MEYER ( 1989), “Lasting Improvement in Manufacturing Performance: In Search K. of a New Theory,” INSEAD Working Paper, Fountainbleau, France. FIEGENBAUM, AND A. KARNANI ( 1991), “Output Flexibility: A Comparative Advantage for Small Firms,” A. Strategic Management Journal, 12, 2, 101-I 14. FINE, C. H. AND A. C. HAX (1985), “Manufacturing Strategy: A Methodology and Illustration,” Interfaces, 15, 6, 28-46.
STATEGY, FLEXIBILITY,
AND PERFORMANCE RELATIONSHIPS
231
FISHER,M. L., A. JAIN, AND J. P. MACDUFFIE( 1995), “Beyond Black” in Redesigning the Firm, B. Kogut and E. Bowman (eds.), Oxford University Press,Oxford, UK. Briefings on this article are provided in
Harvard Business Review ( 1994)) 72, 6, 13- 14.
FORD,J. D. ANDD. A. SCHELLENBERG ( 1982), “Conceptual Issues Linkage in the Assessment of ofOrganizational Performance,” Academy of Management Review, 7, 1,49-58. FRAZELLE, H. ( 1986), “Flexibility: A Strategic Responsein Changing Times,” Industrial Engineering, 18, E. 3, 17-20. GAIMON, C. AND V. SINGHAL( 1992), “Flexibility and the Choice of Manufacturing Facilities Under Short Product Life Cycles,” European Journal of Operational Research, 60, 2, 21 l-223. GERWIN,D. ( 1987), “An Agenda for Research on the Flexibility of Manufacturing Process,” Infernational
Journal of Operations and Production Management, I, I, 38-49.
GERWIN,D. ( 1989), “Manufacturing Flexibility in the CAM Era,” Business Horizons, 32, I, 78-84. ( 1992), “Management of Flexible Automation in the Auto Industry” in Management of R&D and Engineering, D. F. Kocaoglu (ed.), North Holland-Elsevier Science Publishers B. V., Amsterdam, The Netherlands 202-2 17. ( 1993), “Manufacturing Flexibility: A Strategic Perspective,” Management Science, 39, 4, 395-4 10. GIFFI, C., A. ROTH, AND G. SEAL ( 1990), Competing in World-Class Manufacturing: America’s 21st Century Chnllenge, BusinessOne Irwin, Homewood, IL. GOLDHAR,J. AND M. JELINEK( 1983) “Plan for Economies of Scope,” Harvard Business Review, 6 I, 6, 141148. GUPTA,A. AND V. GOVINDARAJAN (1984), “Business Unit Strategy, Managerial Characteristics and Business Unit Effectivenessat Strategy Implementation,” Academy af Management Journal, 27, 1,25-4 I. GUPTA, Y. AND S. GOYAL ( 1989) “Flexibility of Manufacturing Systems: Concepts and Measurements,”
European Journal of Operational Research, 43, 2, I 19-135.
-AND
International
( 1992), “Flexibility Tradeoffs in a Flexible Manufacturing Systems:A Simulation Study,”
Journal of Production Research, 30, 3, 527-557.
-AND -AND-
S. C. LONIAL ( 1994), “The Congruence between Manufacturing Strategy and BusinessStrategy: A Predictor of Organizational Performance,” Working Paper, College of Business,University of Colorado at Denver, Denver, CO. AND T. M. SOMERS 1992), “The Measurement of Manufacturing Flexibility,” European Journal a/ (
Operational Research, 60, 2, I66- 182.
( 1993), “Factory Automation and Integration of BusinessFunctions,” Journal
ufacturing Systems, 12, I, 15-23.
of Man-
GUSTAVSSON, ( 1984), “Flexibility and Productivity in Complex Production Processes,” S. International Journal
af Production Research, 22, 5, 80 l-808.
HAMBRICK,D. C. ( 1984), “Taxonomic Approach to Studying Strategy: Some Conceptual and Methodological Issues,” Journal of Management, 10, I, 27-4 I. -, 1.C. MACMILLAN,AND D. L. DAY ( 1982), “Strategic Attributes and Performance in the BCG MatrixA PIMS-Based Analysis of Industrial Products Businesses,”Academy of Management Journal, 25, 3, 510-531. HAYES,R. H. ( 1981), “Why JapaneseFactories Work,” Harvard Business Review, 59, 4, 56-66. AND S. C. WHEELWRIGHT 1984), Restoring Our Competitive Edge, John Wiley and Sons, New York. ( HOFER, C. W. AND D. S~HENDEL (1978), Strategy Formulation: Analytic Concepts, West Publishing, New York. HUTCHINSON, G. K. AND D. SINHA ( 1989), “A Quantification of the Value of Flexibility,” Journal of Manufacturing Systems, 8, 1, 47-57.
HYUN, J. AND B. AHN ( 1990), “Flexibility Revisited: Review, Unifying Frameworks,and StrategicImplications,” Mimeo, Korean Advanced Institute of Science and Technology, Seoul, Korea. JAIKUMAR,R. ( 1986), “Postindustrial Manufacturing,” Harvard Business Review, 64, 6, 69-76. KANTER, R. M. AND D. BRINKERHOFF ( 1981), “Organizational Performance: Recent Developments in Measurement,” Annual Review of Sociology, 7, 4, 322-349. KEKRE,S. (ZND SRINIVASAN 1990), “Broader Product Line: A Necessity to Achieve Success?“Management K. ( Science, 36, 10, 1216-1231. KERLINGER,F. L. AND E. J. PEDHAZUR 1973), Multiple Regression in Behavioral Research, Holt, Rinehart ( and Winston, New York. KRAFCIK,J. ( 1988), “Triumph of the Lean Production System,” Sloan Management Review, 30, I, 4 I-52. KUSIAK,A. ( 1985), “Flexible Manufacturing System:A Structural Approach,” International JournalofProduction
Research, 23, 6, 1057-1073.
MACDU~E, J. ( 199I ), “Beyond Mass Production: Flexible Production Systemsand Manufacturing Performance in the World Auto Industry,” Ph.D. Dissertation, Sloan School of Management, MIT, Cambridge, MA. MAGAL, S. R., H. H. CARR, AND H. J. WATSON(1988), “Critical SuccessFactors for Information Center Managers,” MIS Quarterly, 12, 3, 444-46 1.
232
YASH P. GUPTA AND TONI M. SOMERS
San MAJCHRZAK,A. ( 1988), The Human Side of Factory Automation, Jossey-Bass, Francisco, CA. MANDELBAUM, M. ( ]978), “Flexibility in Decision Making: An Exploration and Unification,” Unpublished Ph.D. Thesis, Department of Industrial Engineering, University of Toronto, Toronto, Ontario, Canada. MARSCHAK, AND R. NELSON(1962), “Flexibility, Uncertainty, and Economic Theory,” Metroeconomica, T. 14, I, 42-58. MASUYAMA,A. ( 1983), “Idea and Practice of Flexible Manufacturing System of Toyota,” Proceedings of 7th International Conference on Production Research, Windsor, Ontario, 584-590. MILES, R. E. AND C. C. SNOW( 1978), Organizational Strategy, Structure, and Process, McGraw-Hill, New York. MILGROM, P. AND J. ROBERTS 1990), “The Economics of Modern Manufacturing Technology, Strategy, and ( Organization,” The American Economic Review, 80, 3, 5 I l-528. MILLER, J. G. AND A. V. ROTH ( 1987), “Manufacturing Strategies,” Executive Summary of the 1987 North American Manufacturing Futures Survey, Boston University, Boston, MA. AND ( 1994), “A Taxonomy of Manufacturing Strategy,” Management Science, 40, 3, 285304. NAMBOODIRI, K., L. F. CARTER, N. ANDH. M. BLALOCK 1975), AppliedMultivariate Analysis and Experimental ( Design, McGraw-Hill, New York. NASH, M. ( 1983), Managing Organizational Performance, Jossey-Bass, Francisco CA. San NUNNALLY,J. ( 1978), Psychometric Theory, McGraw Hill, New York. PHILLIPS, W. ( 198I ), “AssessingMeasurement Error in Key Informant Reports: A Methodological Note on L. Organizational Analysis in Marketing,” Journal of Marketing Research, 18, 4, 395-4 15. PORTER, ( 1980), Competitive Strategy, Free Press,New York. M. PRIMROSE, ( 1988), “The Effect of ATM Investment on Costing Systems,” Journal of Cost Management,fir P.
Manufacturing Industries, 2, 2, 27-30.
RICHARDSON, R., A. J. TAYLOR, AND J. R. M. GORDON( 1985). “A Strategic Approach to Evaluating P. Manufacturing Performance,” Interfaces, 15, 6, 15-27. SCHROEDER, G., J. C. ANDERSON, R. AND G. CLEVELAND 1986), “The Content of Manufacturing Strategy,” ( Journal of Operations Management, 6, 4, 405-416. SETHI, A. K. AND S. P. SETHI ( 1990), “Flexibility in Manufacturing: A Survey,” International Journal q/
Flexible Manufacturing Systems, 2, 4, 289-328. Journal of Production and Op-
SLACK,N. ( 1987), “The Flexibility of Manufacturing Systems,” International
erations Management, 7, 4, 35-45.
SON,Y. K. AND C. S. PARK( 1987), “Economic Measure of Productivity, Quality, and Flexibility in Advanced Manufacturing Systems,” Journal of Manufacturing Systems, 6, 3, 193-206. SRINIVASAN, ( 1985), “Alternative Measure of System Effectiveness:Associations and Implications,” MIS A.
Quarterly, 9, 3, 243-253.
STALK,G. AND T. HOUT ( 1990), Competing against Time, The Free Press,New York. STECKE, AND I. KIM ( 1989), “Performance Evaluation of Pooled Machines of Unequal Sizes:Unbalancing K. versus Balancing,” European Journal of Operational Research, 42, I, 22-38. STIGLER,G. ( 1939), “Production and Distribution in the Short Run,” Journal of Political Economy, 47, 3, 305-327. SUAREZ, M. CUSUMANO, C. FINE( 1992), “An Empirical Study of Manufacturing Flexibility in PrintedF., AND Circuit Board Assembly,” Working Paper 74-92, International Center For Researchon the Management of Technology, Sloan School of Management, MIT, Cambridge, MA. SWAMIDASS, M. AND W. T. NEWELL( 1987), “Manufacturing Strategy, Environmental Uncertainty and P. Performance: A Path Analytic Model,” Management Science, 33, 4, 509-524. TAYMAZ, E., ( 1989), “Types of Flexibility in a Single-Machine Production System,” International Journal of Production Research, 27, I I, 1891-1899. TOMBAK,M. M. ( 1988), “The Importance of Flexibility in Manufacturing,” Wharton/PIMS ResearchCenter Working Paper, No. 89-002, Philadelphia, PA. AND A. DE MEYER ( 1988), “Flexibility and FMS: An Empirical Analysis,” IEEE Transactions on Engineering Management, 35, 2, 101-107. UPTON,D. M. ( 1994), “The Management of Manufacturing Flexibility,” California Management Review, 36, 1, 72-89. -AND M. M. BARASH 1988), “A Grammatical Approach to Routing Flexibility in Large Manufacturing ( Systems,” Journal of Manufacturing Systems, 7, 3, 209-22 I. VENKA+BAMAN,N. ( 1989), “Strategic Orientation of Business Enterprises: The Construct, Dimensionality, and Measurement,” Management Science, 35, 8, 942-962. AND J. H. GRANT ( 1986), “Construct Measurement in Strategy Research:A Critique and Proposal,” Academy of Management Review, 1 I, ml-86. AND V. RAMANUJA~~1986), “Measurement of BusinessPerformance in Strategy Research:A Com( parison of Approaches,” Academy of Management Review, 11, 4, 80 l-8 14.
STATEGY, FLEXIBILITY, -
AND PERFORMANCE RELATIONSHIPS
233
AND ( 1987), “Measurement of BusinessEconomic Performance: An Examination of Method Convergence,” Journal of Management, 13, 1, 109- 122. VICKERYS. K., C. DROGE,AND R. E. MARKLAND ( 1993), “Production Competence and BusinessStrategy: Do They Affect BusinessPerformance,” Decision Sciences, 24, 2, 435-455. WALTON,R. AND G. SUSMAN 1987), “People Policies for the New Machines,” Harvard Business Review, 66, (
2,98-106.
YAO, D. ( 1985) “Material and Information Flows in Flexible Manufacturing Systems,” Material Flow. 2, 2 and 3, 143-149. ZAMMUTO, R. AND E. O’CONNER( 1992), “Gaining Advanced Manufacturing Technologies’ Benefits: The Role of Organization Design and Culture,” Academy of Management Review, 17, 4, 701-728.
doc_425995944.pdf
The report on topics like manufacturing flexibility, Organizational Performance, path analysis technique
PRODUCTION
AND OPERATIONS MANAGEMENT Vol. 5. No. 3. Fall 1996 PIinlPd in u.SA.
BUSINESS STRATEGY, MANUFACTURING FLEXIBILITY, AND ORGANIZATIONAL PERFORMANCE RELATIONSHIPS: A PATH ANALYSIS APPROACH*
YASH P. GUPTA
AND
TONI
M. SOMERS
Graduate School of Business Administration, University of Colorado at Denver, Denver, Colorado 802 17, USA Department of Finance and Business Economics, School of Business Administration, Wayne State University, Detroit, Michigan 48202, USA
It has been argued in the literature that business strategy and manufacturing flexibility independently affect the performance of an organization. However, no empirical examination of the interrelationship among thesethree constructs has been performed. In this paper, basedon a field study of 269 firms in the manufacturing industry, the identified constructs have been used to test a theoretical model using path analysis techniques. Our re+sults indicate that business strategy contributes both directly and indirectly to organizational performance. The findings provide evidenceof direct effectsof(i) businessstrategyon manufacturing flexibility and (ii) manufacturing flexibility on organizational performance. (MANUFACTURING FLEXIBILITY; BUSINESS STRATEGY; ORGANIZATIONAL PERFORMANCE; PATH MODEL)
1. Introduction
Manufacturing flexibility is one of the most difficult goals for organizations to achieve. Concomitantly, evidence suggeststhat the focus of competition in global markets is shifting from quality and service toward flexibility (Ferdows and De Meyer 1989). In Europe, for example, extensive programs are being initiated to enhance manufacturing flexibility. Theseinclude implementation of advanced manufacturing technology, decrease of cycle times through lead time and setup time reductions (Giffi, Roth, and Seal 1990). Manufacturing flexibility is a multidimensional concept (in this paper manufacturing flexibility and flexibility are used interchangeably). It ensures that the manufacturing process is both cost-efficient and -effective in that it can produce customized products without sacrificing either objective. As setup time decreases, small-batch production can be as economical as large-scale manufacturing, enabling an organization to change its competitive strategy from emphasizing economies of scale to emphasizing economies of scope(Goldhar and Jelinek 1983). Flexibility can be used both as an adaptive response to environmental uncertainty and to proactively create market uncertainties for competition (Gupta and Goyal 1989; Gerwin 1993). For example, in 1981 Honda exploited
* Received December 1993; revised December 1994; accepted April 1995. 204 lO59-1478/96/0503/204$I .25
Copyright 0 1996. Productmn and Operations Management Society
STATEGY,
FLEXIBILITY,
AND
PERFORMANCE
RELATIONSHIPS
205
the benefits of flexibility in inducing customers to expect more frequent changes from the motorcycle industry. In their battle with Yamaha for supremacy in the industry, over a period of 18 months, Honda introduced or replaced 113 models, effectively turning over its entire product line twice. Interestingly, Yamaha managed to complete only 37 changes during the same period, thus creating the consumer perception that Yamaha motorcycles were old, redundant, and unattractive compared to Honda’s. This forced Yamaha to withdraw its challenge (Stalk and Hout 1990). According to Sethi and Sethi ( 1990, p. 295 ) “manufacturing flexibility clearly has major implications for a firm’s competitive strength. This significant role of manufacturing flexibility makes it a part of the firm’s strategy.” Moreover, flexibility cannot be bought; it must be planned and managed (Gustavsson 1984). Despite the widespread acceptance of flexibility’s role in enhancing our competitive position, the management of flexibility remains poorly understood in industry (Cox 1979; Miller and Roth 1987; Tombak and De Meyer 1988; Zammuto and O’Connor 1992). Manufacturing flexibility does not receive adequate attention at the time of decision making for investment in manufacturing technology (Adler 1988), nor does it receive adequate recognition in the implementation phase (Jaikumar 1986). Slack ( 1987) in a study of 10 manufacturing companies observed that managers had a partial, rather than comprehensive, view of manufacturing ?lexibility. Managers focused on machine rather than system flexibility [using Buzacott’s ( 1982) distinction], frequently limiting themselves to a particular type of resource. While it is very tempting to think in this fashion, this strategy may result in a serious mistake. Machine level flexibility alone (e.g., versatile numerical control (NC) machines) does not adequately ensure a competitive edge and will elevate competitiveness only if the added advantage of flexibility in the management of the system (e.g., alternate routing policy) is present. In the same study, Slack ( 1987) found that managers sought to limit the need for flexibility by pursuing three broad strategies: (i) by limiting product range and discouraging frequent product modifications; (ii) by pursuing make-to-stock rather than make-to-order; and (iii) by matching market segmentation with segmentation of the production system, thus reducing the product range. On the research front, the scope of research on manufacturing flexibility has remained quite narrow. Recent literature has focused on defining types of flexibility and identifying systemsthat exhibit one or more of these (for example, seeBrowne et al. 1984;Chattejee, Cohen, Maxwell, and Miller 1984; Carter 1986; Gerwin 1987; Son and Park 1987; Brill and Mandelbaum 1989; Hyun and Ahn 1990). Few researchers have examined measurement issuesfor various types of flexibility (for example, seeMasuyama 1983; Chung and Chen 1990; Dixon 1992; Gupta and Goyal 1992; Gupta and Somers 1992). There is little agreement on how to define flexibility, how to achieve flexibility, or what are the costsand benefits of more, or less,flexibility. For example, some researchers,have viewed flexibility primarily in terms of programmable machines and capabilities for mixing models in production (Taymaz 1989). Others have viewed this only in terms of versatility of people and skills (Walton and Susman 1987). Additionally, some studiestreat flexibility and flexible manufacturing systems( FMS) as equivalent concepts. In actuality, they are not the same. FMS is one method of acquiring flexibility. Other avenues include workers with broad skills, flexible production management techniques, and the development of networks of dependable suppliers. Researchershave also implicitly or explicitly assumed that more flexibility is always better (notable exceptions are Cusumano 1988; Tombak 1988). Flexibility may actually make the players worse off (for example, see Gaimon and Singhal 1992) in some situations. On the basis of the above discussion, it is apparent that achieving manufacturing flexibility is a critical source of competitive advantage for many organizations. Upton ( 1994) while concurring with this view highlighted that:
206
CEOS
YASH P. GUPTA AND TONI M. SOMERS know this, managers know it, and shop floor operators know it. However, the exhortation heard time and time again to “go forth and be flexible” is hollow and meaningless. Managers find themselves unable to express exactly what it is that needs to be improved, without some more precise way of defining the direction of improvement. Quite often, management needs to identify multiple types of flexibility and split them up so they can be prioritized, measured and improved, each by appropriate mechanisms.
The literature on manufacturing flexibility is fragmented and does not provide answers to the basic question of manufacturing organizations: Given that flexibility is a multidimensional concept and that the need for each dimension may vary, under what level would a given flexibility be appropriate in improving organizational performance? As stated earlier, more flexibility in itself is not necessarily useful. The appropriate degree of flexibility dependson what a firm is aiming to achieve in terms of its products, behavior of its competitors, product-demand characteristics, and other factors (Suarez, Cusumano, and Fine 1992). We deal with the above question by investigating how business strategy and dimensions of manufacturing flexibility are related to organizational performance. 1.1. Motivation for Research Our analysis is motivated in part by the recent work of Get-win ( 1993) who reviewed the literature concerning the strategic aspectsof manufacturing flexibility. He established a researchagenda for the area. First, research is needed to determine the extent of which manufacturing flexibility has an impact on a company’s performance. A few investigators have reported relationships between specific types of flexibility and performance. Kekre and Srinivasan ( 1990), for example, investigated the positive and negative effects of product line breadth (in this paper we equated this to processflexibility or mix flexibility as some researchershave termed it) on firm performance. Their results provided some empirical support for the benefits of process flexibility by demonstrating that a broader product line is associatedwith significant market share benefits and increases in firms’ profitability. We recognize that at the conceptual level, equating of product line to process flexibility and then taking the relationship between product line breadth and performance as evidence of a direct relationship between flexibility and performance may be too aggressive.At the operational level, though incomplete, breadth of product line may be an acceptable measure of process flexibility. The breadth of the product line will have positive effectson market share and profitability only to the extent that manufacturing, marketing, and engineering contribute capabilities that make the products in the broader line more desirable to the customer than the competitions’ products. Fiegenbaum and Karnani ( 1991) analyzed data on 83 industries to study the differences between small and large firms for volume flexibility. They concluded that small firms tend to show more volume flexibility than large firms and that small firms are able to trade cost inefficiency with volume flexibility to increase their profits. Tombak ( 1988) pursued the question of whether flexibility affectsa firm’s performance by using a sample of 1,445 businessunits drawn from profit impact of market strategy (PIMS) database.He examined the relationship between manufacturing flexibility and a firm’s performance in the growth and mature phasesof the product life cycle. He found that flexibility was an important explanatory variable for a firm’s performance. Tombak and De Meyer ( 1988) acknowledged the multidimensional nature of flexibility by arguing that firms planning to introduce FMS should be concerned with both “mix flexibility” and the flexibility needed to accommodate the variance in inputs to the production process. Recently, Suarez, Cusumano, and Fine ( 1992) conducted a field study using 3 1 printedcircuit board (PGB) assembly organizations. They concluded that increase in mix and new product flexibility does not lead to higher costs or lower quality. These studies tend to view flexibility as a unidimensional concept. Jaikumar ( 1986) , for example, implicitly refers to flexibility as the ability of a system to produce a wide variety of parts. Using
STATEGY,
FLEXIBILITY,
AND
PERFORMANCE
RELATIONSHIPS
207
Browne et al.% ( 1984) taxonomy, this flexibility can be termed as “mix flexibility.” Yet this is merely one of the different types of flexibility available to a firm. Fiegenbaum and Karnani ( 1991) consider only volume flexibility. On the other hand, Suarez, Cusumano, and Fine ( 1992) did examine the multidimensional nature of flexibility. However, the special characteristics of the PCB industry and the relatively small size of the sample may prohibit us from generalizing their results to other industries. More specifically, these studies did not examine how other dimensions of flexibility are related to performance, nor did they consider business strategy, which in turn is hypothesized to directly affect performance (Swamidass and Newell 1987) and flexibility (Jaikumar 1986). 2. Background In this section we review the literature relevant to each of the constructs used in this paper. 2.1. Business Strategy Business strategy can be viewed as a part of the widely accepted hierarchy of strategy suggested several researchersincluding Hofer and Schendel ( 1978) and Fine and Hax by ( 1985). This hierarchy can be visualized to have three levels: corporate-level strategy, business-levelstrategy, and functional-level strategy. Corporate-level strategy formulation in conglomerate diversified firms is mainly characterized by the consideration of scope and resourcedeployments. At the businesslevel, the scopeand boundaries of each business unit (SBU) and the operational links with corporate strategy are specified. The basis on which the business unit will achieve and maintain a competitive advantage within its industry is also established.At the functional level [e.g., marketing strategy,manufacturing strategy, and research and development (R&D) strategy], the objectives are to support the desired business level strategy in a manner that will provide a competitive advantage and to determine how the functional level strategieswill complement each other. Recently, Venkatraman ( 1989) categorized the literature on business strategy (henceforth termed asstrategy) measurement approachesinto three types: ( 1) narrative approach, (2) classificatory approach, and ( 3) comparative approach. The narrative approach is basedon a premise that strategy of an organization is unique and should only be described in its holistic and contextual form ( Andrews 1971). The implication of this approach is that strategy can (and should) be best described verbally. Any attempt to develop a measurement schemewill be incomplete. This view of strategy measurement may have a role in the conceptual development; it has limited use for testing theories (Venkatraman 1989). The classificatory approach attempts to uncover underlying dimensions of strategy that are generalizableacrossseveralfirms. This attempt has led to an operational definition in terms of strategy typologies basedon conceptual classifications (for example, seeHofer and Schendel 1978; Miles and Snow 1978; Porter 1980) and strategy taxonomies based on empirical classifications (for example, seeHambrick 1984). The comparative approach attempts to identify and measure key traits of strategy constructs. Consequently, the focus is lesson categorization into one cell of the typology or taxonomy but on measuring the differencesalong a set of characteristicsthat collectively describe the strategy construct (for example, seeVenkatraman 1989). According to Venkatraman and Grant ( 1986) , most of the strategy construct measures are either nominal scaleswith questionable measurement properties or multi-item scales whose measurement properties have not been systematically assessed.Venkatraman ( 1989)) using the comparative approach, developed the theoretical underpinnings of the strategy construct for business strategy by proposing six dimensions of strategy: aggressiveness, analysis,defensiveness, futurity, proactiveness,and riskiness. This dimensionality
208
YASH
P. GUPTA
AND TONI
M. SOMERS
was derived and tested by treating the individual dimensions as a building block. Venkatraman ( 1989) suggestedthat his measure for strategy could be used to test theoretical relationships. In this paper, we have used the measures of strategy provided by Venkatraman ( 1989). Definitions of the six strategy dimensions are as follows: Aggressiveness: This dimension reflects the posture adopted by an organization in allocating its resourcesfor improving market positions at a relatively faster rate than the competitors in its chosen market. Analysis: This dimension refers to the tendency of an organization to search deeper for the roots of problems and to generate the best possible solutions alternatives. It also includes the extent to which an organization usesappropriate management systemssuch as information and control systemsand managerial reward systems. Defensiveness: This dimension captures the defensive behavior of an organization through the extent to which the organization employs cost reduction and efficiency seeking methods. Futurity: This dimension reflects temporal considerations embedded in key strategic decisions, in terms of relative emphasis of effectivenessconsiderations versus efficiency considerations. Emphasis on basic research, for example, is most likely to have longer term focus than application-oriented research programs that reflect shorter term focus. This aspectis operationalized by emphasizing salesforecasting and customer preferences as well as tracking of environmental trends. Proactiveness: This dimension reflects proactive behavior about participation in emerging industries, continuous search for market opportunities, and experimentation with potential responsesto changing environmental trends. Riskiness: This dimension captures the extent of riskiness in various resourceallocation decisions as well as choice of products and markets. 2.2. Manufacturing Flexibility A number of classifications of manufacturing flexibility and measuresare given in the literature (Mandelbaum 1978; Kusiak 1985; Frazelle 1986; Upton and Barash 1988; Gupta and Goyal 1989; Hutchinson and Sinha 1989; Taymaz 1989). A most thorough classification was proposed by Sethi and Sethi ( 1990) who identified 11 different flexibility dimensions. Gupta and Somers ( 1992) developed an instrument to measure manufacturing flexibility and conducted an empirical study to validate the dimensions of flexibility identified by Sethi and Sethi ( 1990). Their study revealed that 11 factors of Sethi and Sethi can be collapsed into 9 measuresof flexibility. To our knowledge, no study has identified various dimensions of flexibility. Using a field study drawn from a large sample of firms, we chose to use the measuresof flexibility suggestedby Gupta and Somers. These measuresare as follows: Machineflexibility deals with the variety of operations that the machine can perform without incurring high costs or expending a prohibitive amount of time in switching from one operation to another. Machine flexibility allows small batch sizes.This yields lower inventory costs, higher machine utilizations, the ability to produce complex parts, and improved product quality. Material-handlingflexibility is defined as the ability of material-handling systems to move different part types effectively through the manufacturing facility. This includes loading and unloading of parts, intermachine transportation, and storage of parts under various conditions of the manufacturing facility. In the end, material-handling flexibility may increase machine availability and reduce throughput times. Processflexibility is defined as the ability of a manufacturing system to produce a set of part types without major setups, something referred to as mix flexibility by Gerwin ( 1987) and Carter ( 1986). Process flexibility is useful in reducing batch sizes and, in
STATEGY,
FLEXIBILITY,
AND
PERFORMANCE
RELATIONSHIPS
209
turn, inventory costs. Becauseit allows machines to be shared, it minimizes the need for duplicate machines. Routingflexibility refers to the ability of a manufacturing system to produce a part by alternate routes through the system. The purpose of routing flexibility is to continue to produce a given set of part types, albeit at a lower rate in the event of unexpected machine breakdown. It allows for efficient scheduling of parts through improved balancing of machine loads. Volume flexibility is the ability of a manufacturing system to be operated profitably at different overall output levels, thus allowing the factory to adjust production within a wide range. Program flexibility is the ability of the system to run virtually unattended for a long enough period. Program flexibility reduces the throughput time through reducing setup times, improving inspection and gauging, and better fixtures and tools. Product andproductionflexibility is the universe of part types that the manufacturing system can produce without adding major equipment and the easewith which new parts can be added or substituted for existing parts, i.e., the easewith which the current part mix can be changed at relatively low cost in a short period. This type of flexibility is dependent on several factors: variety and versatility of available machines, flexibility of material-handling systems,and the factory information and control systems. Marketflexibility can be defined as the easewith which the manufacturing system can adapt to changing market environment. It allows the firm to respond to changeswithout seriously affecting the business and to enable the firm to outmaneuver its less flexible competitors in exploiting new business opportunities. Expansion and marketflexibility is the extent of overall effort needed to increase the capacity and capability of a manufacturing system when required. This flexibility may help shorten implementation time and reduce cost for new products, variations of existing products, or added capacity. 2.3. Organizational Performance Severalauthors have emphasized the importance of the performance concept and have provided prescriptions for improving organizational performance (for example, seeNash 1983). Yet the debate on issues of terminology, levels of analysis, and conceptual basis for assessment performance continues to rage in the academic community (Ford and of Schellenberg 1982). Some researchershave expressedfrustration with the lack of agreement on basic terminology and definition ( Kanter and Brinkerhoff 1981) . Venkatraman and Ramanujam ( 1986) suggest that business performance is a subset of the overall concept of organizational effectiveness.They argue that in its narrowest sense,business performance is associatedwith simple outcome based financial indicators (referred to as financial performance) “that are assumedto reflect the fulfillment of the economic goals of the firm.” The financial performance has been most widely used to determine organizational health of a firm. Typical indicators include return on investment, return on sales,and return on equity. A broader conceptualization of businessperformance includes emphasis on indicators of operational performance (i.e., nonfinancial) in addition to financial indicators. Under this conceptualization, measuressuch as market share, new product introduction, product quality, and market effectiveness might be considered within the domain of businessperformance. According to Venkatraman and Ramanujam ( 1986, p. 804) “the inclusion of performance indicators takes us beyond the black-box approach that seemsto characterize the exclusive use of financial indicators and focuses on those key operational successfactors that might lead to financial performance.” The sources of data to measure financial performance, operational performance, or both can be primary (e.g., data collected directly from organizations) or secondary (e.g., data from publicly available records). However, Venkatraman and Ramanujam ( 1987)
210
in
YASH
P. GUPTA
AND TONI
M. SOMERS
their study of 207 senior-level managers showed that managers tend to be less biased in their assessments their organizational performance than researchershave tended to of give them credit for. They argue that the perceptual data can be employed as acceptable operationalizations of organizational performance. Similarly, they demonstrate that a positive and statistically significant association exists between primary and secondary businessperformance data.
3. Hypotheses
In this section we justify the relationship between business strategy, manufacturing flexibility, and organizational performance. Our adaptation of Gerwin’s ( 1993) conceptual model, shown in Figure 1, makes explicit the expected links among the three variables: businessstrategy, manufacturing flexibility and organizational performance. For the purpose of simplicity, the diagram shown in Figure 1 does not show all direct and indirect paths expressedby the model. In this model, business strategy is equated with various types of manufacturing flexibility and posited as a primary influence on an organization’s performance. Specifically, the recursive causal chain shown in Figure 1 hypothesizes that business strategy will trigger the development and implementation of manufacturing flexibility dimensions. The introduction of manufacturing flexibility enhances the organization’s performance. Gerwin’s conceptual framework describes the effects of uncertainty on manufacturing strategy, flexibility, and firm performance. In this study we have used business strategy instead of manufacturing strategy. Our reasoning is based on the fact that manufacturing flexibility is not only an element of manufacturing strategy but also related to marketing and R&D strategies(Hyun and Ahn 1990; Sethi and Sethi 1990). According to Milgrom and Roberts ( 1990), it may be unprofitable for a firm to enhance manufacturing flexibility without changing its marketing strategy or to alter its marketing approach without adequate manufacturing flexibility, and yet, it may be highly profitable to do both together. One can, for example, think of several situations where a
BUSINESS STRATEGY
b
MRNUFACTUFlIN6 FLEXIBILITY
A
4
b
ORGANIZATIONflL PERFORMANCE FINANCIAL 0 GROWTH
FIGURE 1. Conceptual Framework.
STATEGY,
FLEXIBILITY,
AND
PERFORMANCE
RELATIONSHIPS
211
firm’s manufacturing function has done a superior job of developing process,expansion, and volume dimensions of flexibility. However, the marketing function did not take advantage of the resulting opportunities, because of lack of ability or, astonishingly, because marketing remained unaware of manufacturing’s capabilities (Cleveland, Schroeder, and Anderson 1989; Vickery, J%oge,and Markland 1993). This situation may result in manufacturing flexibility having a limited impact on a firm’s performance. Successful manufacturing firms exploit complementarities that exist between various functions, such as marketing, production, engineering, and organizational variables. The exploitation of complementarities among functions is usually dictated by a coherent businessstrategy of the firm (Milgrom and Roberts 1990). Moreover, several researchers have arguedthat an organization should develop a manufacturing strategythat is consistent with and linked to its business strategy (e.g., seeHayes and Wheelwright 1984). A study by Richardson, Taylor, and Gordon ( 1985) provided further evidence of the need to better understand the linkage between corporate strategy and manufacturing strategy.A field study of 64 Canadian electronics companies led the researchers conclude to that companies with a strong match between their business mission and manufacturing tasks are more profitable. They argue that the way a firm competes in its markets is a key element in determining the corporate mission. Performance will be suboptimal unless a proper congruence is achieved between the corporate mission and manufacturing strategy. Cleveland, Schroeder, and Anderson ( 1989) used a sample of six hard goods manufacturers to examine the relationship between businessstrategy, production process,production competence, and business performance. Production competence was defined as a variable rather than a fixed attribute and measuredfor how well manufacturing strengths and weaknessescomplement the priorities of the business strategy. The study showed that production competence is linked with business performance. Swamidassand Newell ( 1987), using data gathered from 35 manufacturers, studied the relationship between environmental uncertainty, manufacturing strategy,and business performance. They defined manufacturing strategy asconsisting of two types of variables: (i) content variables, such as flexibility and (ii) process variables, such as the role of manufacturing managers in strategic decision making ( RMMSDM). From this they concluded that: (a) greater flexibility leads to better performance; (b) RMMSDM is a function of environmental uncertainty and higher levels of RMMSPM result in improved performance; and (c) an organization may be better able to cope with high uncertainties by increasing manufacturing flexibility and maintaining and ensuring the RMMSDM . Schroeder, Anderson, and Cleveland ( 1986) advocated that manufacturing strategy is a processin which business strategy determines manufacturing mission and distinctive competence. By interacting between adjacent levels of the strategy hierarchy, this process determines manufacturing objectives and leads to the formulation of manufacturing policies. Recently, Gupta and Lonial ( 1994), using data from 175 manufacturing firms concluded that the linkage between manufacturing strategiesand business strategiesis a significant predictor of organizational performance. The above discussion indicates that manufacturing flexibility is one of the elements of manufacturing strategy. The congruence between manufacturing strategy and business strategy leads to increased organizational performance (Figure 1) . Therefore, it may be appropriate to examine the relationship between business strategy and manufacturing flexibility. 3.1. Manufacturing Flexibility and Business Strategy Hayes ( 1981) suggestedthat when organizations produce more products, they experience a higher level of demand variability. Similarly, the supply variability will be dependent on the number of unique parts sourced, the number of suppliers used, and the
212
YASH
P. GUPTA
AND
TONI
M. SOMERS
introduction of new or alternate materials. An organization that operates in an industry that is facing an increasingly rapid pace of technological evolution and, consequently, a shortening product life cycle and higher product turnover would be faced with additional variability. Higher levels of labor turnover, absenteeism, and low equipment reliability are additional sources of variability. To deal with these sources of variability, the organization must increase its manufacturing flexibility. When a firm decides to compete in the high end of markets, generally requiring more customized products, it would clearly need more process flexibility. Similarly, should a firm decide to compete in several related industries, it may require more processflexibility than when competing in a single industry. When competitors of an organization are constantly introducing many new and improved products embracing a wide variety of features, almost all dimensions of flexibility would be required to meet competitors’ challenge ( Krafcik 1988;MacDuffie 1991; Suarez, Cusumano, and Fine 1992). The above discussion suggeststhat: l HI : Business strategy has direct effects on the adoption of manufacturing flexibility dimensions. HI would be expected to hold whether business strategy’s effect on manufacturing flexibility results from management’s intention to pursue a competitive advantage or from a strategic necessity. The literature supports the position that business strategy incorporates the required flexibility (Slack 1987; Swamidassand Newell 1987). 3.2. Manufacturing Flexibility and Organizational Performance Kekre and Srinivasan ( 1990) empirically examined the relationship between breadth of product line (process flexibility or mix flexibility as termed by several authors) and market success.They found that broader product lines result in larger market share and profitability and that it does not seem to be associatedwith higher costs. Suarez, Cusumano, and Fine ( 1992), by examining the works of Stigler ( 1939) and Marschak and Nelson ( 1962), suggestedthat in some casesvolume flexibility may be associatedwith higher costs and/or lower quality levels. They advocated that a plant that can shrink and expand production volume and can still keep its costslow and quality high would reap higher levels of performance. Another dimension of flexibility that was studied by Suarez, Cusumano, and Fine ( 1992) is called new product flexibility (termed as product flexibility in this paper). They defined this as the ability to create new products quickly. Drawing from the works of Clark and Fujimoto ( 1991)) Suarez, Cusumano, and Fine ( 1992) argued that as technology advances rapidly and customers become more sophisticated, rapid product introduction can give firms a real competitive advantage,i.e., the organizational performance could be significantly better than its competitors. Moreover, successfulfirms that pursue different businessstrategiesmay score higher on one performance dimension or another. Firms pursuing aggressivestrategiesand higher levels of flexibility, for example, may be expected to score more highly on growth performance than on financial performance. This causal logic suggeststhe following hypotheses: l HZ: Manufacturing flexibility dimensions have direct effects on an organization’s growth ( financial ) performance. H2 provides a test of the relationship between manufacturing flexibility dimensions and an organization’s performance, and similar to previous studies (Kekre and Srinivasan 1990, Fiegenbaum and Kamani 1991), the hypothesis suggestsa direct association between the variables. l H3: Besidesdirect effects, business strategy also indirectly affects an organization’s growth (financial) performance through its effect on manufacturing flexibility dimension.
STATEGY,
FLEXIBILITY,
AND
PERFORMANCE
RELATIONSHIPS
213
H3 is formulated basedon the recent work of Gerwin ( 1993) who through his conceptual model, identified the need for an applied research agenda involving manufacturing flexibility and other related variables: Specifically, research attention needsto be devoted to studying the interrelationship among business strategy, manufacturing flexibility, and organizational performance. This hypothesis proposes that business strategy’s positive effecton an organization’s performance can be enhanced by linking it with the appropriate manufacturing flexibility dimension. From H3 it follows that (i) if the required manufacturing flexibility dimension is inappropriate to the businessstrategy,the indirect effects may be negative, even when the direct effectsof business strategy are positive and (ii) if the business strategy is inappropriate and direct effects on organizational performance are negative, then indirect effects through manufacturing flexibility could exaggerateor equalize businessstrategy’sdirect effects.To our knowledge, this hypothesis has not been examined in prior statistical research in manufacturing flexibility. In this study, path analysis provides a holistic approach for evaluating H, , Hz, and H3 in an integrative framework (Figure 2).
4. Method
In this section we provide an exposition of sample and data collection procedures and measurement of variables used in the study.
4.1. Sample and Data Collection
A total of 1,600 questionnaires were mailed nationwide to five types of U.S. manufacturing firms with 250 or more employees obtained from a mailing list directory: precision machinery, electrical and electronics, industrial machinery, metal products, and automobile and auto part firms. These types of firms were selected based on standard industrial codes (SIC’S).We chose these types of firms (industry segments) becausethey have acknowledged adoption of advanced manufacturing processes. After 2 weeksof initial mailing, a reminder letter with a fresh copy of the questionnaire was mailed to the nonrespondents. As a result of a carefully designed questionnaire, administered according to Dillman’s suggestions( 1978), the sample was believed to be characteristic of firms in their respective industries. A single respondent per organization, typically holding the title Of CEO, President, or Vice Resident of Manufacturing completed the survey instrument. The manufacturing firms were located nationwide. Of 1,600 questionnaires distributed, 279 were completed and returned, representing a responserate of 17.4%.This is a typical response rate for studies of this kind (Magal, Carr, and Watson 1988). The exclusion of questionnaires with incomplete data resulted in a final sample of 269 firms. A copy of the questionnaire is available from the authors. The rationale for selecting the five types of manufacturing firms was to create a diverse sample from which it would be easier to generalize the results. We tried to sample randomly within each industry; however, nonresponse, even with follow-ups, resulted in unequal response among the firm types. The sample is composed of large firms, with 56.4% of the firms having annual sales between 25 1 and 1,000 million dollars and 4 1.1% having 50 l- 1,000 total employees. Table 1 presents a profile of the sample. The extent of differences between respondents and nonrespondents can seldom be determined. Sometimes, however, one can make very limited checks for differences. In light of the low response rate, tests for nonresponse bias were considered necessary.To assess whether the firms included in the study were representative of those firms in the database,a comparison for selecteddemographic characteristics suggests that there were no significant differencesbetween the groups along this setof characteristics.The similarity of respondents versus nonrespondents was assessed using the Kolmorgorov-Smimov (K-S) test. None of the test statistics were statistically significant at p < 0.05.
214
YASH P. GUPTA AND TONI M. SOMERS
S,
S,
EQdoeenousVg&MgS
Ft _ExpemioalMarlret Flexibility F2- Material Handling Flexibility F3- Routing Flexibility F4- Machine Flexibility F5- Market Flexibility P6- Prcduct/Pmduction Fhibility F7- Process Flexibility Fg- Programming Flexibility F9- Volume Flexibility Pp Financial Perfmnance PO cimti Performance -
St - Aggres8ivems.s S2- Analysis S3- Defensivenes S4- Futurity s5- Pmmivencss S6- Riskiness Thesingle-headed representcausal behveen armws the tii these variables. simplicity, all linksare For not shown the in diagram, werecognized and examinedthis but toexist are in Paper.
2. FIGURE Proposed Path Analytic Model Of The BusinessStrategy, Manufacturing Flexibility, and Organizational Performance Relationship.
4.2. Operational Measures of the Variables In this section we will describe items used in measuring the variables used in our study.
4.2.1 BUSINESSSTRATEGY. A 2 l-item scale, extracted from the previous work on business strategiesby Venkatraman ( 1989), was used to measure six types of manufacturing strategies: ( 1) aggressiveness,(2) analysis, (3) defensiveness, (4) futurity, (5) proactiveness, and (6) riskiness. Respondents were asked to indicate the extent of their agreement or disagreement with each of the items using a five-point scale ranging from l-strongly agreeto 5-strongly disagree. An overall summary measure for each of the six
STATEGY, FLEXIBILITY,
AND PERFORMANCE RELATIONSHIPS TABLE 1
Sample Profile
215
Title of the respondents, % Vice President or Director of Manufacturing Plant Manager, CEO or COO Vice President or Director of Technology/Engineering Vice President of Planning Vice President or Director of Other Areas Vice President or Director of Marketing Vice President or Director of R&D Range of salesof the organizational unit, % Less than $25 million 25-50 m 51-100 m 101-250 m 251-500 m 501-1000 m Over 1000 m Total number of employees, % Under 10 1l-50 51-100 101-250 25 l-500 501-1,000 1,oo l-5,000 5,001-10,000 Over 10,000 Businesscategory, % Precision machinery Electrical and electronics Industrial machinery Metal products Automobile and auto parts Others * n = 269.
59.3 14.6 13.8 4.8 4.8 1.9 0.7 4.1 6.7 15.3 13.4 22.4 34.0 4.1 0.0 1.1 1.1 4.1 21.3 41.1 25.7 3.0 2.6 15.8 29.9 17.9 17.9 14.9 3.6
business strategieswas calculated as the average of the items in their respective strategic dimension. The scalehas been found to have high internal consistency reliability in prior empirical studies (Venkatraman 1989). The original group of items was subjected to reliability analysis, and five items were removed resulting in a 24-item scale. Appendix 1 provides the items that were removed. The six business strategies, represented as endogenous variables in the path model, their empirical indicators, and Cronbach’s coefficient alpha are portrayed in Table 2. 4.2.2. MANUFACTURING FLEXIBILITY. In a previous study, the authors developed an instrument for measuring and analyzing manufacturing flexibility (Gupta and Somers 1992): Thirty-four items affecting manufacturing flexibility were identified from the literature, and a preliminary instrument was created to measure them (see Appendix 2 for a list of the items). Respondents were asked to indicate the extent of their agreement or disagreement with each of the items on a five-point Likert-type scale ranging from l-strongly agreeto 5-strongly disagree.The final instrument, after modification, contained 21 items and had an alpha coefficient of 0.888. The large number of items required that they be condensed into a meaningful and manageabledata set. This was accomplished using principal components factor analysis. This technique assumesthat related variables describe a common underlying manufac-
216
YASH P. GUPTA AND TONI M. SOMERS TABLE 2
Business Strategy Constructs, Their Indicators, and Standardized Alpha Coeficient
Indicators We sacrifice short-term profitability to gain market share. We have cut prices to increase our market share. A strong preference for setting prices below the competition. Seeking market share positions at the expense of cash flow and nrofitabilitv. We emphasize effective coordination among functions (e.g., operations and marketing). When confronted with a major decision, we usually try to develop thorough analyses. Planning techniques (PIMS models, portfolio models). Regular manpower planning and performance appraisal. Outputs of management information and control systems. Cost control systems for monitoring performance. Emphasis on product quality through the use of systems such as quality circles. Project management techniques (e.g., PERT/CPM). Our criteria for resource allocation generally reflects shortterm considerations. Forecasting key indicators of operations. Formal tracking of significant general trends. “What-if” analvsis of critical issues. We constantly seek to indentify new opportunities closely related to our operation. We are usually the first ones to introduce new products or services in our markets. We are constantly on the lookout for business units we that can acquire. Operations in the later stages of the life cycle are strategically eliminated (i.e., liquidated, diversted). We adopt a rather conservative view when making major decisions. Approval of new projects on a “stage-by-stage” basis rather than with a “blanket approval.” Operations have generally followed the “tried-and-true” paths. A strong tendency to support projects where the expected rates of return are certain.
Constructs
Alpha Coefficients
0.66
+G+
Defensiveness
0.59
0.58
0.65
0.61
Wing flexibility dimension. Table 3 presents details of this analysis. As Table 3 shows, the 21 items yielded a 9 multi-item factor structure of manufacturing flexibility that explained 72% of the variance in the data. Each factor represents a unique flexibility dimension whose meaning is based on the items that make it up. These nine factors and Cronbach’s coefficient alphas are portrayed in Table 4. The reliability coefficients of each factor ranged from 0.70 to 0.89. Furthermore, the 21-item instrument had a criterionrelated validity of 0.7301, which correlated significantly with a separatemeasureof overall flexibility. In this study, the firm’s overall manufacturing flexibility was assessed asking by respondents to rate “To what extent is your manufacturing flexible?” This criterion question employed a five-point scale: 1 = highly inflexible, 2 = generally inflexible, 3 = neutral, 4 = generally flexible, and 5 = highly flexible. The correlation of each of the nine factors with the criterion ranged from 0.55 to 0.68.
STATEGY, FLEXIBILITY,
AND PERFORMANCE RELATIONSHIPS
217
TABLE 3 Rotated Factor Matrix of 21-Item Instrument to Measure Man@facturinx Flexibility Factor Number Item J I K L A D Z AA BB Q P DD cc E F S G X Y H G N 1 0.76114 0.12991 0.69472 0.63127 0.61855 0.53061 0.32194 0.79452 0.77289 0.57358 0.72849 0.72985 0.82217 0.7985 I 0.36621 0.83683 0.6854 0.83929 0.52378 0.86753 0.53824 0.58261 0.32588 0.8613 0.32385 0.31314 2 3 4 5 6 7 8 9
Factor Name: 1. Expansion/Market Flexibility; 2. Material Handling Rexibility; 3. Routing Flexibility; 4. Machine Flexibility; 5. Market Flexibility; 6. Product/Production Flexibility; 7. Process FIexibility; 8. Programming Flexibility; 9. Volume Flexibility. Note: Factor loadings provided by principal components analysis (varimax rotation) accounted for 72% of the total variance in the data. EigenvaIues exceeded 1.Ofor all factors except factors 8 and 9 which were 0.999.
Nunnally ( 1978) suggestedthat Cronbach’s alpha value of 0.7 is considered adequate for internal consistency reliability. For relatively new scales,however, Nunnally suggested that an alpha value of 0.6 is acceptable. Srinivasan ( 1985), on the other hand, argued that alpha value of 0.5 or higher should be considered sufficient in exploratory research. Moreover, alpha is a function of the number of items in the composite and tends to be conservative. On the basis of these observations, we decided those alpha values over 0.55 were acceptable.
4.2.3. ORGANIZATIONAL PERFORMANCE. An organization’s performance is a multifaceted construct that defies measurement by a single number. Venkatraman ( 1989) used two dimensions, growth performance and financial performance, to measure organizational performance. Each dimension of performance was measured by multiple items adapted from Gupta and Govindarajan ( 1984) (see Table 5 ). Two dimensions of performance, representedas endogenous variables, were measured by asking respondents to indicate their organization’s level of performance relative to their competition with respectto each of the six criteria. The responseoptions, anchored on a five-point Likerttype scale,ranged from ( 1) much worse than competition to ( 5) much better than competition. To assess reliability of these criteria, we calculated Cronbach’s ( 1951) coefthe ficient alpha. Two performance dimensions, suggestedby Venkatraman ( 1989), were calculated for each firm by averaging the responsesfor their respective indicators. In Table 5, we present the performance variables.
218
YASH P. GUPTA AND TONI M. SOMERS TABLE 4
Manufacturing Flexibility, Their Indicators, and Standardized Alpha CoefJicients
Item
Indicators
Constructs
Alpha Coefficients
.I. Time that may be required to double the output of the system is likely to be extremely low. I. Cost of doubling the output of the system is likely to be extremely low. K. The capacity (e.g., output per unit time) of the system can be increased when needed with ease. L. The capability (e.g., quality) of the system can be increased when needed with ease. A. Time required to introduce new products is extremely low. D. Time required to add a unit of production capacity is extremely low. Z. The ability of material-handling systems to move different part types for proper positioning and processing through the manufacturing facility is extremely high. AA. The ratio of number of paths the material-handling systems can support to the total number of paths is very high. BB. The material-handling system can link every machine to everv other machine.
Q. Decrease in throughput because of a machine breakdown is extremely low. P. Cost of the production lost as a result of expediting a preemptive order is extremely low. DD. The number of different operations that a typical machine can perform without requiring a prohibitive cost in switching from one operation to another is very high. CC. The number of different operations that a typical machine can preform without requiring a prohibitive time in switching from one operation to another is very high. E. Shortage cost of finished F. Cost of delay in meeting customer orders is extremely low. S. Number of new parts introduced per year is very high. G. Size of the universe of parts the manufacturing system is capable of producing without adding major capital equipment is extremely large. X. Changeover cost between known production tasks within the current production program is extremely low. Y. The ratio of the waiting costs of processed parts and the total output is extremely low. H. The manufacturing system is capable of running virtually unattended during the second and third shift.
Routing flexibility
Machine flexibility
0.85
0.81
0.89
0.85
0.70
profitably is extremely high.
Volume flexibility
STATEGY,
FLEXIBILITY,
AND
PERFORMANCE
RELATIONSHIPS
219
TABLE 5 Financial and Growth Performance Indicators and Standardized Alpha Coefficient Alpha Coefficients
Indicators
Constructs
5. Model Results In this study, path analysis is the technique we use to explore our propositions about the sequential relationship among business strategy, flexibility, and organizational performance with complex direct and indirect effects,for example, such as those in H3. The most direct conversion of the conceptual model (see Figure 1) involved the specification of a path model (see Figure 2) that would allow for the examination of both the direct and indirect effects of strategy and flexibility on organizational performance. It must be pointed out that Figure 2 is simplified to show only a few of the direct and indirect strategy-organizational performance connections, and it does not show all the paths tested in accord with the effects anticipated in our hypotheses. The model is recursive insofar as it is assumedthat reciprocal causation in the form of causal feedback loops does not exist (Kerlinger and Pedhazur 1973). While performance may affect future business strategy and/or flexibility adoption, these effects are best modeled by considering time lags and therefore are outside the scope of this research. Although this assumption may not be totally justified, casual feedback loops have not yet been demonstrated empirically in this literature and would require complex estimation techniques (Cohen and Cohen 1983). In this model, our hypotheses are made more explicit through the construction of an arrow diagram depicting the expected causal sequence. The resulting diagram shows businessstrategyas an exogenousvariable affecting the required manufacturing flexibility and organization’s performance. Manufacturing flexibility and organizational performance are endogenous variables that are explained partly by business strategy, and in the caseof performance, manufacturing flexibility. Variables are classified as exogenous (independent) to the model when they only emit arrows; endogenous (dependent) variables receive arrows. Each arrow is called a path, and paths connecting the variables can be either direct or indirect. Both direct and indirect strategy and performance connections are shown in the diagram, and the paths are drawn in accordance with the effectsanticipated in the hypotheses.Direct paths are those connections between variables comprised of a single pathway. A standardized path coefficient, or beta, indicates the direction (either plus or minus) and magnitude of influence between variables. For example, the standardized path coefficient between manufacturing flexibility and organization’s performance is the standardized beta weight resulting from the prediction of performance from manufacturing flexibility with business strategy already partialled out.
220
YASH
P. GUPTA
AND
TONI
M. SOMERS
Indirect paths are compound pathways (with mediating variables) made up of several direct pathways. We expect there to be an indirect path from business strategy to performance that is comprised of several significant direct paths. To test Hi, the regression results and the standardized path coefficients (pc.) representing the direct effects of business strategy on manufacturing flexibility are shown in Table 6. The effects of business strategy on manufacturing flexibility are indeed quite distinguishing. Two strategies,aggressiveness proactivenesswere significant predictors and of several types of flexibility. On the other hand, the strategies of a business oriented toward futurity or riskiness were not found to have any significant impact on type of manufacturing flexibility. An aggressivestrategy was a significant determinant of all types of manufacturing flexibility. A proactive strategy emerged as a direct significant predictor of expansion and market, material handling, routing, machine, market, programming, and volume flexibility. Four strategies had significant direct effects on expansion and market flexibility: aggressiveness(p.c. = 0.313, p < 0.01); analysis (p.c. = 0.139, p c 0.10); defensiveness(p.c. = -0.149, p < 0.05); and proactiveness (p.c. = 0.133, p -C 0.05). In fact, a defensive strategy was significantly related to only this type of flexibility (p.c. = -0.149, p < 0.05). The negative coefficient suggests with increased that emphasis on cost reduction and efficiency-seeking methods, the lesslikely the firm’s need to increase the capacity and capability of its manufacturing system. An analysis strategy was significantly associated with both expansion and market (p.c. = 0.139, p < 0.10) and product and production flexibility (p.c. = 0.149, p < 0.05). This suggeststhat an increase in a firm’s overall problem-solving posture is accompanied by increases in (i) capacity and capability of its manufacturing system when needed and (ii) part types that the system can produce without requiring additional major equipment. The models differed in their ability to explain variance in type of manufacturing flexibility. An examination of the adjusted R2 statistics in Table 6 indicates that a modest amount of variance in type of flexibility has been explained by businessstrategy. Business strategy as a whole explained from 1 to 17% of the variance in type of manufacturing flexibility. All models, except one, significantly demonstrated that business strategy has an impact on manufacturing flexibility. It would appear, however, that business strategy has ,no direct effect on product and production flexibility (F = 1.38, p = 0.2 195) even though aggressiveness (p.c. = 0.110, p < 0.10) and analysis (p.c. = 0.149, p < 0.10) strategies approached significance at the 0.05 level. The best models were found for
TABLE
6
Direct Effect of Business Strategy on Manufacturing Flexibility
FI Independent Variables Aggressiveness S IAnalysis S2 Defensiveness S3 Futurity S4 Proactiveness S5 Riskiness S6 Adjusted R* F-Ratio (Probability) Expansion and Market 0.3 13’ 0.139*** -0.149;. 0.066 0.133.’ 0.026 0.17 10.16. (0.~) F2 Material Handling 0.13989 0.02 -0.028 0.05 I 0.191. -0.013 0.06 3.75+ (0.0013) F3 F4 Dependent Variable9 F5 F6 Product and Production 0. I lo*** 0.149*** 0.026 -0.055 -0.0 I 5 0.036 0.0 I 1.38 (0.2 195) F-7 F8 F9
Routing 0.165’ 0.127 -0.103 0.058 0.150** -0.006 0.08 4.64; (0.0002)
Machine 0.209; 0.087 -0.083 0.065 0.190* -0.016 0.1 I 6.45* (0.oooo)
Market 0.158’ 0.025 0.004 0.02 0.159. 0.05 I 0.06 3.61* (0.0019)
Process 0.136”’ 0.012 -0.069 0.106 0.099 0.04 I 0.03 2.24” (0.0397)
Programming 0.229. 0.082 -0.064 0.017 0.158’ 0.022 0. I 5.92’ (0.~)
Volume 0.136*’ 0.002 -0.022 0.058 0.236* -0.014 0.08 4.57* (0.0002)
Note:N=269;*p<0.01;**p<0.05;***p<0.10. * Values are standardized regression coefficients.
STATEGY,
FLEXIBILITY,
AND
PERFORMANCE
RELATIONSHIPS
221
predicting expansion and market and machine flexibility. In these models, business strategy accounted for 17 and 1I%, respectively, of the variance. In a path model, the path coefficients not only identify the direct effect of each of the exogenous variables on the appropriate dependent variables, but they can also be used to calculate both the indirect and the total effects of each variable on the respective dependent variables. As seen in Table 7, the total effect is simply the sum of the direct effectsand all the indirect effectsthat occur through intervening variables. The indirect effect of a variable is that which is traceable through its association with other variables. Estimates for indirect effectsare computed by tracing in the path diagram (see Figure 2) from businessstrategy (S l-S6 ) through the nine types of manufacturing flexibility (FlF9) to performance (growth and financial). For example, the indirect effects of aggressivenessstrategy (S 1) are computed according to the “tracing rule” as: 0.03588 = 0.113(0.313) + 0.056(0.139) + 0.065(0.165) + 0.035(0.209) + 0.003(0.158) + -0.011(0.110) + -0.111(0.136) + -0.061(0.229) + 0.033(0.136) It has been suggestedthat having calculated the path coefficients for a just-identified model, path coefficients that do not meet criteria of statistical significance be deleted from the model (referred to as theory-trimming approach). Following the theory one would delete path coefficients that are not statistically significant at a prespecified level of significance. When it is desired to test whether more than one path coefficient within a given equation may be deleted, it is appropriate to test them simultaneously by using the F-test. This study, however, did not apply the theory trimming approach that would reduce the total number of variables to the smallest core of variables that could efficiently predict growth and financial performance. To test H2 and H3, the effects of business strategy and manufacturing flexibility on growth and financial performance were decomposed into direct, indirect, and spurious. Many (5 of 6) of the business strategies were found to have an impact on the firm’s financial performance. Aggressiveness(p.c. = 0.267, p < 0.01 ), analysis (p.c. = 0.183, p < 0.05), defensiveness (p.c. = -0.129, p < O.lO), proactiveness (p.c. = 0.191, p < .O.O and riskiness ( p.c. = -0.2 19,p -C0.0 1) have significant direct effectson financial 1) performance. The absolute value of the standardized path coefficient allows us to determine which strategiesare more important for financial performance. Thus, aggressiveness and riskiness strategiesare the most important predictors of financial performance. The positive coefficientsimply that financial performance increasesasa firm’s strategicposture increases.In contrast, a negative coefficient, for example, that found for riskiness strategy, suggests that financial performance decreasesas the extent of risk taking increases. Only one type of flexibility, processflexibility, significantly affectsfinancial performance (p.c. = -0.111, p c 0. lo), and its effect is modest, at best. Nonetheless, an increase in the capability of a manufacturing system to produce a set of part types without major setup is accompanied by a reduction in financial performance. The overall adjusted R2 for the model is 0.283 (p < 0.01). Surprisingly, when growth performance is considered as the dependent variable, several types of manufacturing flexibility significantly affect it. In contrast, we found that only one businessstrategy, proactiveness, appeared to be significantly related to firms’ growth performance. The growth performance of a firm is influenced by expansion and market (p.c. = 0.242, p < 0.01 ), routing (p.c. = 0.143, p -C 0.05), product and production (p.c. = -0.128, p c 0.05 ), process ( p.c. = -0.098, p < 0.10) and volume flexibility (p.c. = 0.162, p -C 0.01) as well as by a proactive business strategy (p.c. = 0.253, p -C0.0 1). Overall, almost 20% of the variance in growth performance is explained by the model (Figure 3 ) .
TABLE
7
Direct, Indirect, and Total E#ect of Business Strategy and Manufacturing Flexibility on Organizational Performance
Dependent Variable: Growth Performance
Dependent Variable: Financial Performance Total 0.302* 0.203* -0.146** -0.008 0.220; -0.224* 0.335* 0.178* -0.124** -0.010 0.218* 0.243* 0.101*** 0.032 -0.050 0.033 0.095*** -0.003 0.191* 0.059 -0.123** 0.098*** 0.348* -0.048 0.242* 0.166* 0.242* 0.090 0.027 -0.073 0.065 0.253* -0.045 0.113*** 0.053 0.033 -0.025 0.022 -0.002 -0.002 0.019 spurious
Independent Variable
Direct
Indirect
r
Indirect 0.056 0.05 I 0.014 -0.050 -0.007 0.028 Total
Direct
spurious
r
0.247* 0.1 lo*** -0.109*** 0.048 0.341* -0.020
Sl s2 s3 s4 S5 S6
0.267: 0.183+* -0.129*** -0.014 0.191* -0.219*
0.035 0.020 -0.017 0.006 0.029 -0.005
Fl
0.113
-
-
0.110*** 0.014 0.143** 0.036 0.069 0.010 0.107*** 0.046 0.028
0.352: 0.024 0.250* 0.082 0.097***
F2 F3 F4 F5 0.056 0.065 0.035 0.003 0.014 0.143** 0.036 0.069 0.066 0.197* 0.069 0.058
0.056 0.065 0.035 0.003
0.122** 0.262* 0.104*** 0.06 1
-
F6 F7
-0.011 -0.111*** -0.011 -0.111*** -0.042 -0.131** -0.053 -0.242:
-
-0.128** -0.098’::
-0.128** -0.098***
0.007 0.044
-0.121** -0.054
-0.058 0.162* -
F8 F9
-0.06 1 0.033 -0.06 1 0.033 -0.066 0.068
0.283 8.062 (0.000) -0.127** 0.101***
-0.058 0.162*
-0.047 0.100***
-0.105*** 0.262* 0.191 5.236 (0.0000)
Aggressiveness Analysis Defensiveness Futurity Proactiveness Riskiness Expansion and market flexibility Material-handling flexibility Routing flexibility Machine flexibility Market flexibility Product and Production flexibility Process flexibility Programming flexibility Volume flexibility Adjusted RZ F-Ratio (Probability) Ratio of Correlations duplicated within 0.10
13115
12115
Note: n = 269, * p -c 0.01; ** p < 0.05; *** p i 0.10.
STATEGY,
Direct
FLEXIBILITY,
AND
PERFORMANCE
RELATIONSHIPS
223
Effects of BusinessStrategy on Manufacturing Flexibility and Direct Effects of BusinessStrategy and Manufacturing Flexibility on Financial And Growth Petionnance:
sl,
s2,
s3,
.%,S6
.-b
F4
Sl, ss
F5 F8
F9
Sl,
s2
b b
Fa
Sl
Fl
F7, F9
SS
-b
Fl, F3, Fa,
BusinessStrategy Indirect Effects on Growth Performance‘Ihmugb Manufacturing Flexibility :
s1-----b
ss /
FI .-b
PC
Endo=nous SI - Expansion/Market Flexibility
Sa - Materiel Handling Ss - Routing Flexibility SI _ Machine Flexibility Ss - Market Flexibiity Ss - Product/Production Flexibility
VarMIss
Flexibility
FI - Expansion/Market Flexibility Fz - Material Handling Flexibility R - Routing Flexibility F4 - Machine Flexibility Fs - Market Flexibility F6 - Pmduct/Production Flexibility FT - Process Flexibility F8 - Programming Flexibility Fs _ Volume Flexibility PP. Financial Performance PO - Growth Perfcmnance
FIGURE 3. Summary of Significant Paths from Figure 2 Based on Standardized Regression Coefficients.
It was predicted that any relations between business strategy and organizational performance would be mediated by type of manufacturing flexibility. We expect there to be an indirect path from business strategy to growth and financial performance that is comprised of two significant direct paths. The strength of an indirect path is determined by computing the product of the standardized (direct) path coefficients that make up such a compound path. The results in Table 7 demonstrate partial support for H3, which had posited that manufacturing flexibility would play a mediating role between businessstrategy and the organizational performance of firms. The role of manufacturing flexibility asan intervening variable is confirmed by the finding that aggressive proactive business and strategies marginally influence growth performance indirectly through their effects on manufacturing flexibility. Contrary to expectations, none of the business strategies had indirect effects on financial performance through manufacturing flexibility. Business
224
YASH
P. GUPTA
AND
TONI
M. SOMERS
strategy indirect effects on financial through flexibility are quite small. These business strategiesapparently have no bearing on manufacturing flexibility. It would appear then that part of the relationship between businessstrategyand financial performance is indirect through other variables not included in this analysis. Billings and Wroten ( 1978) suggested once path coefficients have been determined, that they should be verified by attempting to recompute the correlation matrix through the calculation of the total effectsbetween the setsof related variables. In Table 7, a comparison of the estimated correlations, found by summing the direct and indirect effects (total effect), with the original correlations ( r) between the independent variables and dependent measuresprovides supporting evidence of the “goodness of fit” of these models. If we used the criterion that the absolute difference between the reproduced and original correlations should not exceed 0.10, our models can duplicate almost all ( 13 of 15) of the original correlations for financial performance and 12 of the 15 correlations involving growth performance (Namboodiri, Carter, and Blalock 1975). Further, the extent of the spurious effectsfound for both models suggests that unless business strategy is considered in these models, the direct effects of manufacturing flexibility on growth and financial performance will be underestimated. For some variables, the spurious effects and their direct effects are of opposite sign, implying that the zero-order correlation coefficient alone would provide a misleading relationship between manufacturing flexibility and firms’ performance (Table 8). A summary of the results reported in this section is given in Figure 3. 6. Discussion The objective of this paper was to determine the relationship between businessstrategy, manufacturing flexibility, and organizational performance. The data used to test our hypotheses and the development of flexibility measuresreported by Gupta and Somers ( 1992) are drawn from the same database.Therefore, findings of this research put them in more of an exploratory rather than confirmatory category. The aggressiveness dimension of business strategy is significantly related to all of the dimensions of manufacturing flexibility identified in this paper (Table 6 ) . Since aggressive organizations tend to sacrificeprofitability and set prices below competition to gain market share even at the expense of cash flow profitability and since manufacturing flexibility allows an organization to increase market share by improving its ability to produce a greater mix of products, change the volume of production, and respond to changing market conditions, it is obvious that the aggressiveness strategy should be significantly
TABLE
Zero-Order FI Expansion and Market FI F2 F3 F4 F5 F6 Fl F8 F9 F2 Material Handling Correlation F3 Matrix F4
8
Flexibility F6 Product and Production n Variables F8 F9
for Manufacturing F5
Routing
Machine
Market
Process
Programming
Volume
0.3697’ 0.2226’ 0.2131* 0.2608* 0.3893* 0.0489 0.2787* 0.3816* 0.1248" 0.1059”’ 0.3579* 0. I506** 0.2041* 0.3524* 0.2505; 0.19541 0.2816* 0.2827* 0.0084 0.3869* 0.2768* 0.1508** 0.2683* 0.2504; 0.3999* 0.3052*
0.2864 0.0637 0.2208’ 0. I954f
0.0287 0.0726 0.2501*
0.0537 0.3747s
0.2933; -
STATEGY,
FLEXIBILITY,
AND
PERFORMANCE
RELATIONSHIPS
225
related to manufacturing flexibility. These results are further supported by the findings reported in Table 7. In this table, the direct effect of aggressivelypursuing market share (aggressiveness)on growth performance is not significant; however, aggressiveness orientation affectsgrowth indirectly through manufacturing flexibility, suggestingthat firms opting for aggressiveness strategy and aiming for an increasing growth trend must enhance manufacturing flexibility. Proactive organizations generally seek new opportunities for business that can be acquired and are generally the first to introduce new products. Organizations pursuing this strategy seemto seekall the flexibility dimensions except processflexibility and product and production flexibility (see Table 6 ) . In addition, the proactiveness strategy also influences growth performance indirectly through manufacturing flexibility. One possible explanation, albeit weak, for the lack of a direct relationship between proactiveness and process and product and production flexibility dimensions is that the newly acquired businessesand their associated products are generally kept independent of the current business units. Thus, the current manufacturing plant may not have a great need for processflexibility. The path analysis results demonstrate that aggressiveness proactiveness strategies and influence growth performance indirectly through manufacturing flexibility (Table 7 ) . These findings suggestthat the linkage between strategy and growth performance is more complex than suggestedby previous researchand thus emphasize the need for additional studies rising multivariate research designs. It is interesting to observe that organizations pursuing a defensivenessstrategy tend to seekvery little manufacturing flexibility. This strategy tends to be reactive and emphasize the implementation of cost reduction and efficiency improvement methods. These characteristics generally do not require high levels of manufacturing flexibility (Giffi, Roth, and Seal 1990). The lack of a significant relationship between analysis and futurity dimensions of business strategy and dimensions of manufacturing flexibility may be attributed to the fact that defensivenessorientation in businessstrategy is strongly related to futurity and analysis (see Venkatraman 1989). Our results show that the aggressiveness,analysis, defensiveness, proactiveness, and riskiness dimensions of business strategy are significantly related to the financial performance of an organization (Table 7). The directions (signs) of all these relationships, except aggressivenessand defensiveness, are consistent with those obtained by Venkatraman ( 1989). Our results on the association between aggressivenessand financial performance, however, are in consonance with Hambrick, Macmillan, and Day’s ( 1982) findings. They suggestedthat market share gains can be achieved without loss of current profitability. One plausible explanation could be that our study deals with the data collected exclusively from manufacturing companies, whereas Venkatraman collected data from a number of industries. In any event, the inconsistencies in results suggest the need for additional field studies in this area. The significant and positive relationship between proactiveness and growth performance is supported by previous research (Venkatraman 1989). Our findings indicate that five dimensions of flexibility out of nine are significantly related to growth performance, whereas only one dimension significantly affectsfinancial performance (Table 7). It is conceivable that since manufacturing flexibility is still a relatively new phenomenon, management may not yet have recognized this linkage between flexibility and financial performance. Moreover, Kekre and Srinivasan ( 1990) suggested that growth performance (market share) should eventually stimulate financial performance of an organization. Expansion and market flexibility have a positive and significant impact on growth performance. This flexibility allows a firm to respond to changes in customers’ tastes, declining product life cycles, and uncertainty in sources of supplies and, thus, enables it
226
YASH
P. GUPTA
AND TONI
M. SOMERS
to become a time-based competitor. These findings are consistent with the literature, i.e., time-based organizations tend to perform significantly better than their competitors who do not pursue time-based strategies (Stalk and Hout 1990). To enhance this flexibility the processof production planning and inventory controls must be integrated with marketing functions, such as product development and market forecasts.Similarly, improved relationships with suppliers and well-developed distribution channels are essential (Sethi and Sethi 1990). The impact of routing flexibility on growth performance suggeststhat organizations in their attempt to enhance this flexibility must employ multipurpose machines, pool identical machines into machine groups (Stecke and Kim 1989)) improve versatility of material handling systemsand system control software (Yao 1985), plan for underutilization of machines, or build redundancy in machines so that the production system can be rescheduled and maintain the overall production rate in caseof a machine breakdown. This also suggests training programs designedto increasemaintenance workers’ that knowledge of the production system to prevent damage and to reroute production may be beneficial (Gerwin 1989). It is interesting to observe that process flexibility impacts negatively on both growth and financial performance. One plausible explanation for this result is that a large number of organizations have implemented automated manufacturing technology with the fuzzy notion to increase process flexibility. However, they have continued to use these technologies under massproduction regimes. Thus, they do not necessarily benefit from the potential increasein processflexibility while incurring the cost of investment in automated technologies (see for example, Bessant 1985; Jaikumar 1986; Majchrzak 1988; Primrose 1988). Another explanation for the negative relationship between processflexibility and performance hasbeen put forward by Gerwin ( 1993) who suggested increasedproduct that variety (process flexibility) leads to complexity and confusion that raisesoverhead costs. Contrary to expectations, product and production flexibility has a negative relationship with growth performance. As noted by Ettlie and Penner-Hahn ( 1994), the true value of this type of flexibility is not realized until the next generation of products is introduced. It is possible that a significant number of our respondents may not have launched a new generation of products since improving this dimension of flexibility. The finding that volume flexibility has a significant and positive relationship with growth performance suggeststhat workers must possess skills that can be used elsewhere when production volume decreases. This, in turn, implies that management must provide cross-functional training to their employees(Gerwin 1989). In addition, volume flexibility can be enhanced by realizing the importance of subcontracting networks and implementing just-in-time concepts. 6.1. Implications The management of manufacturing flexibility has been cited as a critical issue for both manufacturing executives and general managers (Upton 1994). The increasing beliefs in the strategic role of manufacturing in general (Hayes and Wheelwright 1984), along with the level of executive manager participation in our study, suggestthat manufacturing flexibihty will continue to be a critical issue well into the 1990s. This study was driven by a strong desire to provide empirical evidence that could form the basis for the guidelines for managing flexibility. In this article, evidence is provided that manufacturing managers should not increase all dimensions of flexibility in their drive to make improvement in their organization’s performance. Some dimensions of flexibility may have a counterproductive effect,and others may not significantly contribute to their organizational performance goals. Consequently, the investments in flexibilityenhancing mechanisms must be in consonance with the dimension of flexibility that management may be attempting to improve.
STATEGY,
FLEXIBILITY,
AND
PERFORMANCE
RELATIONSHIPS
227
Several dimensions of flexibility (for example, expansion and market and volume) allow an organization to respond to markets expeditiously. Those organizations who know their markets better may not need as much flexibility. On the other hand, less competent organizations who are unable to predict the market changes may favor flexibility. According to Ettlie and Penner-Hahn ( 1994) “these sameorganizations may have difficulty implementing flexibility and other manufacturing innovations.” Improved expansion and market flexibility provide greater freedom to experiment with variations in product designs. This, in turn, will result in escalation in transactions between manufacturing and product design functions and “may lead to greater conflict unless the additional experimentation is prevented from interfering with regular production (Gerwin 1992, p. 2 15) .” Senior executives of an organization must improve working relationships between thesefunctions by implementing mechanismsthat enhance communications and encourage input from manufacturing in product design process (Gupta and Somers 1993) . Gerwin ( 1992) suggested increasedexpansion and market that flexibility may reduce pressure on product engineers to design the products “right” the first time because the inadequacies or defects in product design may be screened and modified when prototypes are being produced or even after, causing escalation in uncertainty for manufacturing function. The above discussion of results suggeststhat the most effective way to increase manufacturing flexibility is to invest heavily in flexible manufacturing capabilities, including technology, organizational systems, and worker training (worker skills are especially important). According to Fisher, Jain, and MacDuffie ( 1994)) the companies that fail in their efforts to increase flexibility do so not because they haven’t bought the right hardware, but because they either don’t understand the importance of worker training or are hampered in their efforts to institute it.
7. Limitations and Future Research
Although this study provides interesting insights into the pattern of relationships among various dimensions of manufacturing flexibility, business strategy, and organizational performance, the results must be interpreted cautiously. It should be pointed out that although the adjusted R2 statistics of 0.283 and 0.191 in Table 7 would be considered acceptable, it must also be concluded that there are other determinants of successthat were not included in our model. These factors may include environmental uncertainty as measuredby Swamidassand Newell ( 1987) or components of manufacturing strategy other than manufacturing flexibility as recognized by Miller and Roth ( 1994) and Gupta and Lonial ( 1994). It would be worthwhile to expand our model in the future studies to include these factors. Similarly, while our results support Hi, in that the adjusted R2 statistics are significant (Table 6)) not all the coefficients in each equation turned out significant. Further, all but three of the adjusted R2 statistics are <O.lO. These results indicate that there are probably a number of other factors affecting the choice of each type of flexibility. More insight on flexibility must come from subsequent studies. Three of the six constructs of business strategy have Cronbach alpha values below 0.6, which Nunnally identified as the boundary for reliability, and three are barely above that. Although the precedents exist in the literature where alpha value of 0.5 and above wasconsideredacceptable,the low value may be attributed to the mixing of manufacturing industries in our sample. A future study may focus on a single manufacturing industry so that the above issue may be resolved. In addition, the single-manufacturing industry study would allow researchersto control confounding market variables that may vary from industry to industry. The study used a single respondent per organization. Seeking responsesfrom a single informant to make judgments on complex organizational characteristics may increase the subjective propensity of single respondents to seekout consistency in their responses
228
YASH P. GUPTA AND TONI M. SOMERS
and increase random measurement error. According to Miller and Roth ( 1994) the random error components may result from the reporting process,knowledge deficiencies, inadequate measures,or some combination of these factors. Campbell ( 1955) identified two criteria for collecting information about a social system: ( 1) the respondents should occupy roles that make them knowledgeable about the issues being researchedand (2 ) they should be willing and able to communicate with the researchers.In our study, the monorespondent problem may have been moderated by the fact that high-ranking respondents tend to be more reliable sourcesof information than their lower ranking counterparts (Phillips 1981) . Moreover, we believe that such a strategy enabled us to achieve a greater and more diverse sample size. Future studies, however, should attempt to avoid this methodological pitfall by pragmatically obtaining multiple sources of information within single organizations. This study has established the relationship between business strategy, manufacturing flexibility, and performance; it would be an important research issue to examine the impact of dimensions of flexibility on quality and productivity. Also, it would be useful to examine how various types of organization structures may have a moderating effect on the need for manufacturing flexibility. 8. Conclusions In this study we have examined the relationship between business strategy, manufacturing flexibility and performance. Our results indicate that businessstrategy contributes both directly and indirectly to organizational performance. The findings provide evidence of direct effectsof(i) businessstrategyon manufacturing flexibility, and (ii) manufacturing flexibility on organizational performance.’
’ We acknowledgethe constructive comments made by two reviewers and the associateeditor on the previous version of this paper.
Appendix 1. Business Strategy Scales
Item Our information systemsprovide support for day-to-day decision making. We have made significant modifications to our manufacturing technologies. We emphasize basic researchto provide us with future competitive edge. Our competitors generally preempt us by expanding capacity ahead of us. Our mode of operations is riskier than our competitor’s operations.
Strategy Scale Analysis Defensiveness Futurity Proactiveness Riskiness
The five items above were removed based on an examination of the item-to-total correlation data, which suggested that these items could be removed from their respective scales.Low correlations for these items suggestedthat they may not share the core of the construct.
Appendix 2. Initial List of 34 Items Measuring Manufacturing
Flexibility
and Their Source
Code A B C
Item Description Time required to introduce new products is extremely low. Cost required to introduce new products is extremely high. Time required to increase or decreaseproduction volume by 20% is extremely low.
Source Sethi and Sethi (1990) Sethi and Sethi ( 1990) Sethi and Sethi (1990)
STATEGY, FLEXIBILITY,
AND PERFORMANCE RELATIONSHIPS
229
Appendix 2. (continued) Item Description Time required to add a unit of production capacity is extremely low. Shortage cost of finished products is extremely low. Cost of delay in meeting customer orders is extremely low. Size of the universe of parts the manufacturing system is capable of producing without adding major capital equipment is extremely large. The manufacturing system is capable of running virtually unattended during the second and third shift. Cost of doubling the output of the system is likely to be extremely low. Time that may be required to double the output of the system is likely to be extremely low. The capacity (e.g., output per unit time) of the system can be increased when needed with ease. The capability (e.g., quality) of the system can be increased when needed with ease. The per unit manufacturing cost is extremely stable over widely varying levels of total production volume. The range of volumes in which the firm can run profitably is extremely low. Average number of possible ways in which a part type can be processedin the system is extremely high. Cost of the production lost as a result of expediting a preemptive order is extremely low. Decreasein throughput becauseof a machine breakdown is extremely low. Time required to switch from one part mix to another is extremely low. Number of new parts introduced per year is very high. Cost required to switch one part mix to another is extremely low. Total incremental value of new products that can be fabricated within the system for a 20% additional cost in new fixtures, tools, and part programs is extremely low. Volume (number of different part types or range of sizesand shapes)of the set of part types that the system can produce without major setups is extremely low. Extent to which product mix can be changed while maintaining efficient production is very narrow. Changeover cost between known production tasks within the current production program is extremely low. The ratio of the total output and the waiting cost of parts processed is extremely low. The ability of material handling system to move different part types for proper positioningand processing through the manufacturing facility is extremely high. The ratio of the number of paths the material handling systemscan support to the total number of paths is very high. The material handling system can link every machine to every other machine. The number of different operations that a typical machine can perform without requiring a prohibitive time in switching from one operation to another is very high. The number of different operations that a typical machine can perform without requiring a prohibitive cost in switching from one operation to another is very high. Source Sethi and Sethi (1990) Abadie et al. (1988) Abadie et al. (1988) Chatterjee et al. (1984) Jaikumar (1986) Carter ( 1986) Carter (1986) Sethi and Sethi (1990) Sethi and Sethi (1990) Falkner (1986) Sethi and Sethi (1990) Chatterjee et al. (1984) Ball ( 1989) Browne et al. (1984); Buzacott ( 1982) Browne et al. ( 1984); Buzacott ( 1982) Jaikumar (1986) Browne et al. (1984); Buzacott ( 1982) Jaikumar (1986) Gerwin (1987) Carter ( 1986) Warnecke and Stienhipler (1982) Son and Park ( 1987) Sethi and Sethi (1990) Sethi and Sethi (1990) Chatterjee et al. (1984) Sethi and Sethi (1990) Sethi and Sethi (1990)
Code
D E F G H. 1 J K L M N 0 P
Q
R S T U V W X Y Z AA BB cc DD
230
YASH P. GUPTA AND TONI M. SOMERS Appendix 2. (continued)
Code EE FF GG HH
Item Description The ratio of the total output and the idle cost of a typical machine for a given period is very low. The number of tools or the number of programs that a typical machine can use is very low. The extent of variations in key dimensional and metallurgical properties of the raw input stock a typical machine can handle is very low. The rate at which a typical machine becomes obsolete when a new product is introduced is very high.
Source Son and Park (1987) Tarondeau ( 1982) Gerwin (1987) Lam (1987)
References
ADLER, P. S. ( 1988), “Managing Flexible Automation,” California Management Review, 30, 2, 34-56. ANDREW&K. R. ( 197 1), The Concept ofcorporate Strategy, Dow Jones-Irwin, New York. BESSANT, ( 1985), “The Integration Barrier: Problems in the Implementation of Advanced Manufacturing J. Technology,” Robotica, 3, 2, 97-103. BILLINGS,R. S., AND S. P. WROTEN( 1978), “Use of Path Analysis in Industrial/Organizational Psychology: Criticisms and Suggestions,” Journal of Applied Psychology, ,63, 6, 677-688. BRILL,P. AND M. MANDELBAUM 1989), “On Measuresof Flexibility in Manufacturing Systems,”International (
Journal qfProduction Research, 27, 5, 747-756.
BROWNE, D. DUBOIS,K. RATHMILL, S. P. SETHI, AND K. E. STECKE(1984), “Classification of Flexible J., Manufacturing Systems,” The FMS Magazine, 2, I, I 14-117. BUZACOTT, A. ( 1982), “The Fundamental Principles of Flexibility in Manufacturing Systems,” Proceedings J. of the 1st International Conference on Flexible Manufacturing Systems, Brighton, UK, (October 2022).
CAMPBELL, T. ( 1955), “The Informant in Quantitative Research,” American Journal of Sociology, 60, 4, D.
339-342.
CARTER,M. F. ( 1986), “Designing Flexibility into Automated Manufacturing Systems,” Proceedings of the
2nd ORSA/TIMS Conference on Flexible Manufacturing Systems: Operations Research Models and Applications, K. E. Steckeand R. Suri (eds.), Elsevier SciencePublishers, Amsterdam, The Netherlands.
CHATTERJEE, M. COHEN,W. MAXWELL,AND L. MILLER ( 1984), “Manufacturing Flexibility: Models and A., Measurements,” Proceedings of the 1st ORSA/TIMS Conference on FMS, Ann Arbor, Ml, 49-64. CHUNG,C. H. AND I. J. CHEN( 1990), “Managing Flexibility of Flexible Manufacturing Systemsfor Competitive Edge” in The Selection and Evaluation ofAdvanced Manufacturing Technologies, M. J. Liberatore (ed.), Springer-Verlag, Berlin, 280-305. CLARK,K. B. AND T. FUJIMOTO 199 I). Product Development Performance: Strategy, Organization, and Man( agement in the World Auto Industry, Harvard BusinessSchool Press,Boston, MA. CLEVELAND, R. G. SCHROEDER, J. C. ANDERSON 1989), “A Theory of Production Competence,” G., AND (
Decision Sciences, 20, 4, 655-668.
COHEN,J. AND P. COHEN( 1983), Applied Multiple Regression/Correlation Analysis for the Behavioral Sciences, Erlbaum, Hillsdale, N.J. COX,T. ( 1979))“Toward the Measurementof Manufacturing Flexibility,” Production and Inventory Management
Journal, 30, 1, 68-72.
CRONBACH, J. ( 1951), “Coefficient Alpha and the Internal Consistency of Tests,” Psychometrika, L.
297-334.
16, 2,
CUSUMANO, ( 1988), “Shifting Economies: Craft Production to the Flexible Factory,” Working Paper 2012M. 88, Sloan School of Management, MIT, Cambridge, MA. DILLMAN, D. ( 1978), Mail and Telephone Surveys: The Total Design Method, John Wiley & Sons,New York. DIXON, J. R. ( 1992), “Measuring Manufacturing Flexibility: An Empirical Investigation,” European Journal of Operational Research, 60, 2, 131-143. ETTLIE,J. E. AND J. D. PENNER-HAHN 1994), “Flexibility Ratios and Manufacturing Strategy,” Management ( Science, 40, 1I, 1444-1454. FERDOWS, AND A. DE MEYER ( 1989), “Lasting Improvement in Manufacturing Performance: In Search K. of a New Theory,” INSEAD Working Paper, Fountainbleau, France. FIEGENBAUM, AND A. KARNANI ( 1991), “Output Flexibility: A Comparative Advantage for Small Firms,” A. Strategic Management Journal, 12, 2, 101-I 14. FINE, C. H. AND A. C. HAX (1985), “Manufacturing Strategy: A Methodology and Illustration,” Interfaces, 15, 6, 28-46.
STATEGY, FLEXIBILITY,
AND PERFORMANCE RELATIONSHIPS
231
FISHER,M. L., A. JAIN, AND J. P. MACDUFFIE( 1995), “Beyond Black” in Redesigning the Firm, B. Kogut and E. Bowman (eds.), Oxford University Press,Oxford, UK. Briefings on this article are provided in
Harvard Business Review ( 1994)) 72, 6, 13- 14.
FORD,J. D. ANDD. A. SCHELLENBERG ( 1982), “Conceptual Issues Linkage in the Assessment of ofOrganizational Performance,” Academy of Management Review, 7, 1,49-58. FRAZELLE, H. ( 1986), “Flexibility: A Strategic Responsein Changing Times,” Industrial Engineering, 18, E. 3, 17-20. GAIMON, C. AND V. SINGHAL( 1992), “Flexibility and the Choice of Manufacturing Facilities Under Short Product Life Cycles,” European Journal of Operational Research, 60, 2, 21 l-223. GERWIN,D. ( 1987), “An Agenda for Research on the Flexibility of Manufacturing Process,” Infernational
Journal of Operations and Production Management, I, I, 38-49.
GERWIN,D. ( 1989), “Manufacturing Flexibility in the CAM Era,” Business Horizons, 32, I, 78-84. ( 1992), “Management of Flexible Automation in the Auto Industry” in Management of R&D and Engineering, D. F. Kocaoglu (ed.), North Holland-Elsevier Science Publishers B. V., Amsterdam, The Netherlands 202-2 17. ( 1993), “Manufacturing Flexibility: A Strategic Perspective,” Management Science, 39, 4, 395-4 10. GIFFI, C., A. ROTH, AND G. SEAL ( 1990), Competing in World-Class Manufacturing: America’s 21st Century Chnllenge, BusinessOne Irwin, Homewood, IL. GOLDHAR,J. AND M. JELINEK( 1983) “Plan for Economies of Scope,” Harvard Business Review, 6 I, 6, 141148. GUPTA,A. AND V. GOVINDARAJAN (1984), “Business Unit Strategy, Managerial Characteristics and Business Unit Effectivenessat Strategy Implementation,” Academy af Management Journal, 27, 1,25-4 I. GUPTA, Y. AND S. GOYAL ( 1989) “Flexibility of Manufacturing Systems: Concepts and Measurements,”
European Journal of Operational Research, 43, 2, I 19-135.
-AND
International
( 1992), “Flexibility Tradeoffs in a Flexible Manufacturing Systems:A Simulation Study,”
Journal of Production Research, 30, 3, 527-557.
-AND -AND-
S. C. LONIAL ( 1994), “The Congruence between Manufacturing Strategy and BusinessStrategy: A Predictor of Organizational Performance,” Working Paper, College of Business,University of Colorado at Denver, Denver, CO. AND T. M. SOMERS 1992), “The Measurement of Manufacturing Flexibility,” European Journal a/ (
Operational Research, 60, 2, I66- 182.
( 1993), “Factory Automation and Integration of BusinessFunctions,” Journal
ufacturing Systems, 12, I, 15-23.
of Man-
GUSTAVSSON, ( 1984), “Flexibility and Productivity in Complex Production Processes,” S. International Journal
af Production Research, 22, 5, 80 l-808.
HAMBRICK,D. C. ( 1984), “Taxonomic Approach to Studying Strategy: Some Conceptual and Methodological Issues,” Journal of Management, 10, I, 27-4 I. -, 1.C. MACMILLAN,AND D. L. DAY ( 1982), “Strategic Attributes and Performance in the BCG MatrixA PIMS-Based Analysis of Industrial Products Businesses,”Academy of Management Journal, 25, 3, 510-531. HAYES,R. H. ( 1981), “Why JapaneseFactories Work,” Harvard Business Review, 59, 4, 56-66. AND S. C. WHEELWRIGHT 1984), Restoring Our Competitive Edge, John Wiley and Sons, New York. ( HOFER, C. W. AND D. S~HENDEL (1978), Strategy Formulation: Analytic Concepts, West Publishing, New York. HUTCHINSON, G. K. AND D. SINHA ( 1989), “A Quantification of the Value of Flexibility,” Journal of Manufacturing Systems, 8, 1, 47-57.
HYUN, J. AND B. AHN ( 1990), “Flexibility Revisited: Review, Unifying Frameworks,and StrategicImplications,” Mimeo, Korean Advanced Institute of Science and Technology, Seoul, Korea. JAIKUMAR,R. ( 1986), “Postindustrial Manufacturing,” Harvard Business Review, 64, 6, 69-76. KANTER, R. M. AND D. BRINKERHOFF ( 1981), “Organizational Performance: Recent Developments in Measurement,” Annual Review of Sociology, 7, 4, 322-349. KEKRE,S. (ZND SRINIVASAN 1990), “Broader Product Line: A Necessity to Achieve Success?“Management K. ( Science, 36, 10, 1216-1231. KERLINGER,F. L. AND E. J. PEDHAZUR 1973), Multiple Regression in Behavioral Research, Holt, Rinehart ( and Winston, New York. KRAFCIK,J. ( 1988), “Triumph of the Lean Production System,” Sloan Management Review, 30, I, 4 I-52. KUSIAK,A. ( 1985), “Flexible Manufacturing System:A Structural Approach,” International JournalofProduction
Research, 23, 6, 1057-1073.
MACDU~E, J. ( 199I ), “Beyond Mass Production: Flexible Production Systemsand Manufacturing Performance in the World Auto Industry,” Ph.D. Dissertation, Sloan School of Management, MIT, Cambridge, MA. MAGAL, S. R., H. H. CARR, AND H. J. WATSON(1988), “Critical SuccessFactors for Information Center Managers,” MIS Quarterly, 12, 3, 444-46 1.
232
YASH P. GUPTA AND TONI M. SOMERS
San MAJCHRZAK,A. ( 1988), The Human Side of Factory Automation, Jossey-Bass, Francisco, CA. MANDELBAUM, M. ( ]978), “Flexibility in Decision Making: An Exploration and Unification,” Unpublished Ph.D. Thesis, Department of Industrial Engineering, University of Toronto, Toronto, Ontario, Canada. MARSCHAK, AND R. NELSON(1962), “Flexibility, Uncertainty, and Economic Theory,” Metroeconomica, T. 14, I, 42-58. MASUYAMA,A. ( 1983), “Idea and Practice of Flexible Manufacturing System of Toyota,” Proceedings of 7th International Conference on Production Research, Windsor, Ontario, 584-590. MILES, R. E. AND C. C. SNOW( 1978), Organizational Strategy, Structure, and Process, McGraw-Hill, New York. MILGROM, P. AND J. ROBERTS 1990), “The Economics of Modern Manufacturing Technology, Strategy, and ( Organization,” The American Economic Review, 80, 3, 5 I l-528. MILLER, J. G. AND A. V. ROTH ( 1987), “Manufacturing Strategies,” Executive Summary of the 1987 North American Manufacturing Futures Survey, Boston University, Boston, MA. AND ( 1994), “A Taxonomy of Manufacturing Strategy,” Management Science, 40, 3, 285304. NAMBOODIRI, K., L. F. CARTER, N. ANDH. M. BLALOCK 1975), AppliedMultivariate Analysis and Experimental ( Design, McGraw-Hill, New York. NASH, M. ( 1983), Managing Organizational Performance, Jossey-Bass, Francisco CA. San NUNNALLY,J. ( 1978), Psychometric Theory, McGraw Hill, New York. PHILLIPS, W. ( 198I ), “AssessingMeasurement Error in Key Informant Reports: A Methodological Note on L. Organizational Analysis in Marketing,” Journal of Marketing Research, 18, 4, 395-4 15. PORTER, ( 1980), Competitive Strategy, Free Press,New York. M. PRIMROSE, ( 1988), “The Effect of ATM Investment on Costing Systems,” Journal of Cost Management,fir P.
Manufacturing Industries, 2, 2, 27-30.
RICHARDSON, R., A. J. TAYLOR, AND J. R. M. GORDON( 1985). “A Strategic Approach to Evaluating P. Manufacturing Performance,” Interfaces, 15, 6, 15-27. SCHROEDER, G., J. C. ANDERSON, R. AND G. CLEVELAND 1986), “The Content of Manufacturing Strategy,” ( Journal of Operations Management, 6, 4, 405-416. SETHI, A. K. AND S. P. SETHI ( 1990), “Flexibility in Manufacturing: A Survey,” International Journal q/
Flexible Manufacturing Systems, 2, 4, 289-328. Journal of Production and Op-
SLACK,N. ( 1987), “The Flexibility of Manufacturing Systems,” International
erations Management, 7, 4, 35-45.
SON,Y. K. AND C. S. PARK( 1987), “Economic Measure of Productivity, Quality, and Flexibility in Advanced Manufacturing Systems,” Journal of Manufacturing Systems, 6, 3, 193-206. SRINIVASAN, ( 1985), “Alternative Measure of System Effectiveness:Associations and Implications,” MIS A.
Quarterly, 9, 3, 243-253.
STALK,G. AND T. HOUT ( 1990), Competing against Time, The Free Press,New York. STECKE, AND I. KIM ( 1989), “Performance Evaluation of Pooled Machines of Unequal Sizes:Unbalancing K. versus Balancing,” European Journal of Operational Research, 42, I, 22-38. STIGLER,G. ( 1939), “Production and Distribution in the Short Run,” Journal of Political Economy, 47, 3, 305-327. SUAREZ, M. CUSUMANO, C. FINE( 1992), “An Empirical Study of Manufacturing Flexibility in PrintedF., AND Circuit Board Assembly,” Working Paper 74-92, International Center For Researchon the Management of Technology, Sloan School of Management, MIT, Cambridge, MA. SWAMIDASS, M. AND W. T. NEWELL( 1987), “Manufacturing Strategy, Environmental Uncertainty and P. Performance: A Path Analytic Model,” Management Science, 33, 4, 509-524. TAYMAZ, E., ( 1989), “Types of Flexibility in a Single-Machine Production System,” International Journal of Production Research, 27, I I, 1891-1899. TOMBAK,M. M. ( 1988), “The Importance of Flexibility in Manufacturing,” Wharton/PIMS ResearchCenter Working Paper, No. 89-002, Philadelphia, PA. AND A. DE MEYER ( 1988), “Flexibility and FMS: An Empirical Analysis,” IEEE Transactions on Engineering Management, 35, 2, 101-107. UPTON,D. M. ( 1994), “The Management of Manufacturing Flexibility,” California Management Review, 36, 1, 72-89. -AND M. M. BARASH 1988), “A Grammatical Approach to Routing Flexibility in Large Manufacturing ( Systems,” Journal of Manufacturing Systems, 7, 3, 209-22 I. VENKA+BAMAN,N. ( 1989), “Strategic Orientation of Business Enterprises: The Construct, Dimensionality, and Measurement,” Management Science, 35, 8, 942-962. AND J. H. GRANT ( 1986), “Construct Measurement in Strategy Research:A Critique and Proposal,” Academy of Management Review, 1 I, ml-86. AND V. RAMANUJA~~1986), “Measurement of BusinessPerformance in Strategy Research:A Com( parison of Approaches,” Academy of Management Review, 11, 4, 80 l-8 14.
STATEGY, FLEXIBILITY, -
AND PERFORMANCE RELATIONSHIPS
233
AND ( 1987), “Measurement of BusinessEconomic Performance: An Examination of Method Convergence,” Journal of Management, 13, 1, 109- 122. VICKERYS. K., C. DROGE,AND R. E. MARKLAND ( 1993), “Production Competence and BusinessStrategy: Do They Affect BusinessPerformance,” Decision Sciences, 24, 2, 435-455. WALTON,R. AND G. SUSMAN 1987), “People Policies for the New Machines,” Harvard Business Review, 66, (
2,98-106.
YAO, D. ( 1985) “Material and Information Flows in Flexible Manufacturing Systems,” Material Flow. 2, 2 and 3, 143-149. ZAMMUTO, R. AND E. O’CONNER( 1992), “Gaining Advanced Manufacturing Technologies’ Benefits: The Role of Organization Design and Culture,” Academy of Management Review, 17, 4, 701-728.
doc_425995944.pdf