ابعاد اندازه گیری تولید انعطاف پذیر
|کد مقاله||سال انتشار||مقاله انگلیسی||ترجمه فارسی||تعداد کلمات|
|3638||2004||16 صفحه PDF||سفارش دهید||10770 کلمه|
Publisher : Elsevier - Science Direct (الزویر - ساینس دایرکت)
Journal : Journal of Operations Management, Volume 22, Issue 2, April 2004, Pages 171–196
Even though many managers and academics have cited flexibility as a key competitive capability, efforts to measure and understand this complex construct continue. Consequently in this paper, we address the issue of manufacturing flexibility measurement, and then use these measures to better understand flexibility. Churchill’s [J. Market. Res. 16 (1979) 64] paradigm is used to develop psychometrically sound measures for six oft-used dimensions of manufacturing flexibility: machine, labor, material handling, mix, new product, and modification. Previous research shows that each of these dimensions, in turn, is comprised of four elements. The resulting 24 scales (6 dimensions×4 elements) demonstrate the desired properties of unidimensionality, reliability, and validity. We show further that the four elements of any given manufacturing flexibility dimension can be grouped into two conceptually separate factors representing “Scope” and “Achievability” of flexible responses. Scope and achievability factor scores can be used to compare a subset of firms with respect to their flexibility choices, and observe the trade-offs firms make both within and across flexibility dimensions. Along with scale development, establishing scope versus achievability relationships between flexibility elements provides a better basis for measuring and creating a holistic understanding of this complex concept.
The potential competitive impact of flexibility is well recognized (Cox, 1989 and De Meyer et al., 1989). However, both managers and academics have noted the lack of appropriate measures for it (e.g., De Toni and Tonchia, 1998, Parker and Wirth, 1999, Koste and Malhotra, 1999 and Beach et al., 2000) as well as the need to better understand the relationships among different types of flexibility (e.g., Parker and Wirth, 1999 and Beach et al., 2000). While efforts have been made to address the measurement gap, good, generalizable measures that span multiple industries are still lacking (Gerwin, 1993), which in turn hinders the effective management of this key capability. This study seeks to address this need by creating generalizable measures for six different dimensions of manufacturing flexibility and then using these measures to further our understanding of manufacturing flexibility. A number of recent studies have measured flexibility objectively, using industry specific measures (e.g., Dixon, 1992 and MacDuffie et al., 1996; Suarez et al., 1995 and Suarez et al., 1996; Upton, 1995 and Upton, 1997). Unfortunately, these measures are not broad-based and are applicable only to the industries within which they were created. Consequently, three other studies have attempted to develop flexibility measures that span multiple industries. Two of these were empirical, but they were exploratory in nature. Gupta and Somers (1992) provided the first effort at scale development. They culled eleven flexibility dimensions and 34 items from the existing literature. Exploratory factor analysis (EFA) resulted in the retention of nine factors, some of which were composites of the original flexibility dimensions. Several of the measures thus developed were single-item measures and the flexibility dimensions lacked a well-defined theoretical domain. A more recent effort by D’Souza and Williams (2000) partially addresses this concern. The authors were consistent in specifying the domain for the volume, variety, process and material handling flexibility dimensions included in their study. However, while prior research indicates that the domain of flexibility consists of more than two elements (e.g., Gupta and Buzacott, 1989; Slack, 1983 and Slack, 1987; Upton, 1994), the authors used only two to assess each flexibility dimension. Single or dual items from the literature were chosen to capture the range and mobility elements of each of them. EFA demonstrated that the items loaded on the two factors as anticipated. While this study advanced the empirical efforts toward scale development, a concern exists with respect to the fact that D’Souza and Williams (2000) operationalized only two elements. This concern is significant, and mandates additional efforts in measuring flexibility. A study by Parker and Wirth (1999) took a different approach. The authors examined the intended purpose of a given flexibility measure and used evaluative criteria to identify “best” measures for several different types of flexibility. They found two types of flexibility lacked good measures and subsequently created models to assess them. While insightful, the underlying assumptions behind the models were fairly restrictive, potentially limiting their use in industry. The above studies focused solely on flexibility, although this concept has been operationalized in a broader context in other empirical research. Pagell and Krause (1999) examined the relationships between operational flexibility, uncertainty, and firm performance. Narasimhan and Das (1999) examined the effect of strategic sourcing on advanced manufacturing technologies (AMT) and on three specific types of flexibility, as well as the relationships among the flexibility dimensions. Brandyberry et al. (1999) also examined the relationship between AMT and market-oriented flexibility. However, since flexibility was not the primary focus of these studies, the operationalization of this construct was fairly limited. Two of the studies combined multiple dimensions of flexibility into one measure (e.g., Pagell and Krause, 1999 and Brandyberry et al., 1999) or focused on limited aspects of the dimensions (e.g., Narasimhan and Das, 1999 measured flexibility for all dimensions with respect to the mobility element only). While these studies provided insight and advanced their respective research fields, they would have been strengthened considerably by a fuller, multi-faceted approach to measuring flexibility, an objective we intend to pursue in this study.
نتیجه گیری انگلیسی
The primary goal of this study was to address the lack of non-industry specific measures for manufacturing flexibility. Given the multi-dimensional complexity associated with this concept, Churchill (1979) paradigm was an appropriate framework for this study. An extensive literature search generated potential scale items that were supplemented by newly proposed ones. The paradigm was subsequently completed for six flexibility dimensions, each of which is comprised of four elements. Factor analysis revealed that each flexibility dimension response consists of scope and achievability components that are conceptually different, though not completely unrelated to one another. These two factors highlight the need to consider both the benefits and penalties associated with flexibility, so that the potential trade-off between these factors can be evaluated. The two factors, scope and achievability, were subsequently used to illustrate that organizational differences do exist with respect to flexibility. These differences can be found within and across flexibility dimensions as well as industries, thereby providing a richer understanding of the flexibility capability of an organization. Rigorous scale development is a time-consuming endeavor, which is also evidenced by the fact that between 1989 and 1996, Hensley (1999) could find only six studies in the operations management literature that utilized and described a formalized, complete scale development process using questionnaire data. Yet, theory development and testing cannot progress without valid theoretically derived constructs. It is within this context that the contribution of this paper to the knowledge base in flexibility must be assessed. Specifically, 1. It provides psychometrically sound measures at the elemental level for six dimensions of flexibility. While Koste and Malhotra (1999) provided the framework within which manufacturing flexibility could be interpreted and undertook the first step in the paradigm of identifying the domain of the constructs, this study actually created the unique set of generalizable measures and scales that several researchers in the field have called for in published studies. In that regard, the current work is both timely and useful. The measures created possess unidimensionality, reliability, convergent validity, discriminant validity, and predictive validity. The demonstration of these properties has been noted as largely lacking in other flexibility measures (O’Leary-Kelly and Vokurka, 1998), potentially creating a situation where erroneous conclusions are reached. Consequently, the availability of these measures will allow the advancement of this field of research. 2. It empirically tests and validates the framework provided in Koste and Malhotra (1999). That study was conceptual in nature, and the empirical testing of the four elements validates this approach to measuring and understanding flexibility. Further analysis can also yield insights into relationships between flexibility dimensions and the testing of flexibility hierarchy as proposed by these and some other authors. 3. The study identifies two underlying factors of flexibility—Scope and Achievability. While not specifically labeled so in prior literature, these two factors highlight the need to consider both the benefits and penalties associated with flexibility, so that the potential trade-off between these factors can be evaluated. As our literature review reveals, prior measures have generally tended to either blend limited aspects of both benefits and penalties into a single scale, or focus on only isolated narrow versions of benefits and penalties. Our constructs along four elements for each dimension capture much greater depth and richness than the single item measures used in other studies such as D’Souza and Williams (2000). Moreover, the scales we report and the two factors of scope and achievability contained within them are also relevant for practitioners, who may need to decide which aspect of a given flexibility dimension is most pertinent to their organization and thus demand appropriate investment of organizational resources. As with any scale development effort of this magnitude and complexity, we recognize the limitations associated with this research. Presence of additional contextual information for each sample firm would have provided greater insight, but was not feasible with such a large-scale data collection effort that spanned multiple industry groups. Consequently, as with any newly developed scales, this research should be replicated and also expanded for other dimensions of interest such as volume flexibility. The measurement properties of the scales developed here should also be validated for a range of other SIC groups and process choices. Such future studies would provide additional insight into and confidence regarding the generalizability of the manufacturing flexibility scales developed here. We are hopeful that future scholars would find these valid, reliable, non-industry specific measures useful in advancing several other avenues of research. For example, the flexibility capability of organizations could be assessed within and across industry groups. To date, these comparisons have been restricted to firms within a single industry (e.g., Dixon, 1992; Upton, 1995 and Upton, 1997). Alternately, researchers could also use these scales to explore the evolution and management of this key capability, and attain a better understanding of how firms make tradeoffs between scope and achievability of each type of flexible response.