استراتژی تولید عمومی: آزمون تجربی از دو پیکربندی نوع شناسی

The need to empirically test and validate typologies and frameworks that are derived deductively has been echoed repeatedly in the operations management literature. This paper reports on an empirical comparison of two configuration-based typologies: the Product–process matrix and the more recent generic manufacturing strategies model. Since there is substantial conceptual overlap between these models, a simultaneous examination provides insights about both models, and in particular, about the value-added of the generic manufacturing strategies model. We examine hypotheses derived from these typologies using data from manufacturing plants in the United States, Italy, United Kingdom, Japan, and Germany; and from the automotive, machinery, and electronics industries. The data were analyzed using multiple analysis of variance (MANOVA) and hierarchical regression techniques. Our results indicate support for the Product–process matrix — lending further strength to a growing base of empirical research on this model. Our findings also provide support for the generic manufacturing strategies model with respect to various measures of cost, cycle time/inventory, quality, and innovation performance. Furthermore, our findings suggest that the generic manufacturing strategies model is a useful augmentation to the Product–process matrix. These findings suggest that the generic manufacturing strategy model has merit but deserves further empirical and theoretical attention.

A growing number of scholars are recognizing the need for empirical validation of operations management theories and propositions that are derived deductively (Adam and Swamidass, 1989, Amoako-Gyampah and Meredith, 1989, Flynn et al., 1990 and Swamidass, 1991). Bozarth and McDermott (1998) take the position that “a typology is not fully developed until it has been empirically validated”. In this study, we compare and contrast two of these typologies: one that has been partially validated empirically — Hayes and Wheelwright, 1979a, Hayes and Wheelwright, 1979b and Hayes and Wheelwright, 1984 Product–process matrix, and another, which has not received empirical scrutiny — Kotha and Orne (1989) proposed generic manufacturing strategy framework. Frequently, theoretical frameworks are validated individually. However, substantial conceptual overlap exists between these two theories. Therefore, it is both interesting and useful to simultaneously examine the similarities and differences between these two theories. Where the two theories overlap, similar predictions should emerge from both of them. Competing or differing hypotheses should also result from the differences between these two theories. These similarities and differences can be evaluated empirically, thus, providing both a test of the predictive capability of each theory as well as a test of the differences between them. The value of this approach is two-fold. First, it provides additional empirical validation of the Product–process matrix while providing an initial empirical validation of Kotha and Orne (1989) generic manufacturing strategies framework. The second potential benefit is derived from the simultaneous examination of the two related theories. This approach can provide useful insights about the value added of one theory over and above the other. In a statistical sense, it is analogous to the partial F-test for a new set of variables (suggested by the generic manufacturing strategies framework) in a baseline model (that represents the Product–process matrix). The rest of the paper is organized as follows. We review the relevant literature in Section 2. In Section 3, we develop the hypotheses that will be empirically tested. The research design and description of the variables is presented in Section 4. The estimation procedures are discussed in Section 5. Finally, in Section 6, we present results, discussions, limitations, and future directions of research.

While several typologies of operations management have been suggested in the literature, many of them have not been systematically validated or evaluated in comparison to one another. In this paper we report a conceptual and empirical comparison of the Hayes and Wheelwright Product–process matrix and the Kotha and Orne generic manufacturing strategy framework. We found that the Product–process matrix has predictive capability with respect to manufacturing performance. We also found that the Kotha and Orne framework is a meaningful extension of the Product–process matrix. The addition of organizational scope to the two dimensions of the Product–process matrix adds explanatory power in predicting a broad array of types of manufacturing performance at the plant level. One implication of this finding for researchers is that organizational scope should receive more attention in the operations management literature, since it has received much fruitful attention in the strategic management literature. The identification of ideal types is of both theoretical and practical significance. Theoretically, this study provides some evidence of the empirical validity of the two typologies studied. It also raises some important questions regarding the continuing validity of these typologies. From a practitioner’s point of view, the identification of recognizable typologies is useful, because it helps the plant manager understand how and why the manufacturing plant can improve it’s performance by re-aligning its products, processes, and scope over time to be more consistent with a particularly relevant strategic type. This study supports the general prediction of the Product–process matrix, that operating on the diagonal of the matrix provides higher performance, particularly in the cost and quality areas. This study adds evidence to a growing body of empirical research supporting Hayes and Wheelwright’s Product–process matrix. While adding a third dimension, organizational scope may enhance their framework, the basic predictions are upheld by our study. The implications for managers of this finding are that they should carefully match product and process choices by operating on the diagonal of the Product–process matrix. If they follow this prescription, they can expect their costs to be lower and quality higher than their competitors. Off diagonal strategies appear to lead to lower performance in these areas. The results of this study are also generally consistent with prior studies of “focused” operations. Perhaps future studies could seek to integrate the findings of these lines of research as well. Just how applicable are the Product–process matrix and the generic manufacturing strategy typologies in today’s diverse and changing manufacturing environment? The results of this study do not speak directly to this question, however, the question is important and we believe that it must be dealt with in the future. The typologies examined in this study are based on conceptualizations of technology that are changing. As manufacturing technologies continue to develop and diffuse into industry, it is clear that the typologies examined in this paper will need to be developed and adapted to reflect the changes occurring in manufacturing. Scholars and managers alike might well ask themselves, “If the majority of manufacturers were using AMT’s (of a specific type), how would the results of this study be affected? How would the typologies of this study have to be adapted to remain valid or even increase their predictive capabilities? What are the effects of specific AMT’s upon these typologies? In particular, future studies should explicitly examine (theoretically and empirically) if either or both of these two models are predictive when AMT’s form the dominant technologies in existence in an industry. Additionally, scholars might consider how to specifically adapt these models to the AMT context. We are of the opinion that questions like these should be driving the current and future research agendas of at least some manufacturing strategy scholars. However, it will be difficult to move forward, without more empirical validation of existing theories, frameworks and models. As we have noted in the introduction to this paper, much work remains to be done in verifying and comparing typologies and theories of operations management. In particular, longitudinal research studies are needed to obtain better evidence of causality for theory testing. We also need to develop more precise measurement methods and universal research approaches to facilitate comparisons across studies. This emerging field of typology development in operations management will benefit from further research aimed not only at testing existing typologies, but developing new and more powerful explanatory theories.