The current issue and full text archive of this journal is available at
                                                                 www.emeraldinsight.com/0144-3577.htm
         IJOPM                                                     Operations management
         28,8                                                        research: evolution and
                                                                     alternative future paths
         710                                                                               Christopher W. Craighead
                                                                     Department of Supply Chain and Information Systems,
         Received February 2007                 Smeal College of Business, The Pennsylvania State University, University Park,
         Revised March 2008
         Accepted April 2008                                                                     Pennsylvania, USA, and
                                                                                                        Jack Meredith
                                                          Babcock Graduate School of Management, Wake Forest University,
                                                                                    Winston-Salem, North Carolina, USA
                                                Abstract
                                                Purpose – This paper aims to investigate the evolution of operations management (OM) research
                                                along two major dimensions from 1977 to 2003 and discusses possible paths for research progression
                                                in the future.
                                                Design/methodology/approach – To identify OM research papers, a careful definition of OM
                                                research was constructed based initially on earlier work and then more precisely extended through
                                                empirical analysis. The research on OM builds on a previous study that took snapshots of OM
                                                research in 1977 and 1987. It then extends and updates it through a content analysis of 593 articles
                                                published in 1995 and 2003 in five journals recognized for publishing OM research.
                                                Findings – The overall results illustrate that OM has evolved from heavily rationalistic, axiomatic
                                                analyses based on artificial reconstructions of reality toward more interpretive analyses based on
                                                natural observations of reality.
                                                Research limitations/implications – As the OM field continues to evolve, it is important to
                                                monitorandreassesspublishedresearchtodiscernitschangingdimensions.Whilethiseffortisnotan
                                                exhaustive review of all OM research and does not consider all relevant journals and years, it does
                                                offer the “big picture” perspective needed for analyzing changing research approaches in the field.
                                                Practical implications – The research provides an analysis of the evolution of knowledge creation
                                                within the field and possible paths for its future development. The practical implications are that as
                                                research becomes more interpretive and observation-based, the findings will have more relevance for
                                                managers and the problems they face.
                                                Originality/value – While several authors have analyzed the OM field relative to select research
                                                methods and journals, this paper provides a broader and more encompassing view of OM research
                                                along two important research dimensions: the researcher’s framework and the source of the data.
                                                Keywords Researchmethods,Operationsmanagement,Functionevaluation
                                                Paper type Literature review
                                                Introduction
         International Journal of Operations &  Research in operations management (OM) has changed dramatically over the years.
         Production Management                  Originally concerned with industrial management methods and procedures for
         Vol. 28 No. 8, 2008                    improving processes (Buffa, 1980, p. 1), the field moved through relatively simple
         pp. 710-726
         qEmeraldGroupPublishingLimited mathematical techniques for independent process improvements such as assembly
         0144-3577
         DOI 10.1108/01443570810888625          line balancing and job shop scheduling to more sophisticated management science
             techniques for optimizing flows, blends, and resource allocations. More recently, there           Operations
             has been a movement toward a more diverse set of empirical (i.e. based on direct sense        management
             experiences or observations) and even interpretive frameworks based on                              research
             surveys/questionnaires, case/field studies, and interviews. Moreover, we are also
             seeing changes in the means of data collection such as the use of postal, e-mail, and
             internet communication, direct personal contact with managers, and even personal
             observation of the unit of analysis such as when a plant or manufacturing cell is                        711
             involved.
                The purposes of this paper are to first track the evolution of OM research as
             published in several top journals and then discuss various alternative paths that may
             definethefutureevolutionofresearchinourfield.Weareparticularlyinterestedintwo
             patterns of OM research:
                (1) the rationalistic versus interpretive orientation of the researcher; and
                (2) whether the researcher desires observational or artificial data for conducting
                    the research.
             Wewill elaborate on these two patterns shortly but we wish to emphasize that we are
             not talking about research methods, though methods naturally embody these two
             patterns of interest.
             Background
             In terms of research in OM, 1980 marked an epoch in the field. Two new journals
             devoted to research solely in OM started publication: International Journal of
             Operations & Production Management (IJOPM) in Europe and Journal of Operations
             Management(JOM)inAmerica.Included in that first issue of JOM were two seminal
             articles written by recognized leaders in the field, Elwood Buffa and Richard Chase.
             They summarized the past history of both research content and process in the field,
             and both called for major changes in research approaches. Buffa (1980) forecast that
             future OM research would move away from mathematical optimization, which we
             would characterize as being a heavily rationalistic treatment of artificial
             reconstructions of the situation of interest. Likewise, Chase (1980) noted the
             prevalence of mathematical/computer modeling, the unsophisticated research designs,
             andthepaucityof“macro-oriented”research.ThenarrownessofOMresearch(“micro”
             in the terms of Chase, 1980) was reemphasized by Miller et al. (1981).
                In 1989, Meredith et al. (1989) defined a “Research Matrix” (Meredith et al., 1989,
             Figure 3, p. 309) intended to more explicitly describe the character of research
             approaches other than just listing various research methods. The axes of the matrix
             consisted of two separate dimensions:
                (1) the researcher’s framework ranging from rational (i.e. highly deductive,
                    axiomatic) to existential (inductive, interpretive); and
                (2) the source of the data ranging from natural (i.e. empirical, directly observed) to
                    artificial (typically hypothetical reconstruction).
             These two dimensions better identify research movement in a field because selected
             research methods (e.g. surveys, mathematical modeling, interviews) inherently
             embodyboththeresearcher’sframeworkaswellashowtheresearcherobtainsdatafor
             analysis. For example, the researcher’s framework may be closer to rational than, say,
   IJOPM    interpretive and thus be at the logical positivist/empiricist level, favoring, for example,
   28,8     experimentation. But to obtain data, the researcher can experiment either through
            direct observation of human subjects during laboratory experiments, or by artificial
            reconstruction using computer simulation. Similarly, a researcher may be dedicated to
            artificial reconstruction for obtaining data, but could use mathematical modeling if she
            or he held a logical positivist/empiricist’s framework or conceptual modeling with an
   712      interpretivist’s framework.
              The Meredith et al. (1989) paper identified a series of alternative research
            approaches in OM and plotted the trends of three OM research journals – JOM,
            Management Science (MS), and Decision Sciences (DS) – during the years 1977 and
            1987 on the axes of the Research Matrix. Almost two-thirds of the papers
            wereclassified in the highly rational (axiomatic) category, almost one-third were in the
            adjacent logical positivist/empiricist category, and 9 percent were in the interpretive
            category. There seemed to be a slight shift toward the existential end of the scale over
            the decade. The great majority (93 percent) of the papers fell in the artificial end of the
            natural-artificial continuum and there was no significant change over the decade.
              In 1993, Neely (1993) used a modification of Chase’s categorization to examine all
            the articles published in IJOPM during the decade of the 1980s. His objective was to see
            if the research content and research processes of European (primarily) OM publications
            had changed substantially over the decade. He found that while the content had
            changedfromafocusonsmall,hardissuestolarger(“macro,”inChase’sterminology),
            softer issues, there had been no discernable change in the research processes. He
            speculated that the reason may have been because research content is often driven by
            the P/OM community and environment, whereas the research process (framework and
            data source) is selected by the individual doing the investigation and reflects personal
            preference and situational needs.
              The previously discussed literature, along with many other analyses
            (Amoako-Gyampah and Meredith, 1989; Scudder and Hill, 1998; Pannirselvam et al.,
            1999; Rungtusanatham et al., 2003; Chopra et al., 2004), has provided valuable insights
            into selected aspects of OM research. However, most of the analyses have been
            narrowly focused in one way or another, such as those that have looked at a particular
            topic (Voss, 2005), a select research method (Malhotra and Grover, 1998; McCutcheon
            and Meredith, 1993), institutional productivity (Malhotra and Kher, 1996; Young et al.,
            1996), or particular journals (Chopra et al., 2004; Pilkington and Fitzgerald, 2006).
            Although all of these have been informative, they have typically centered on topics,
            methods, journals, researchers, schools, or some other limited aspect of the field. There
            is a need to examine OM research in a larger, more philosophical framework along
            multiple dimensions, such as those selected in the Meredith et al. (1989) study: the
            perspective of the researcher and the data source. The purpose of this study is to
            provide such a broader, more all-encompassing analysis of research in the field.
            Research study
            Asnotedearlier,thepurposeofthisresearchistotracktheevolutionofOMresearchwith
            afocusontwodimensionsoftheresearchfromthepasttothepresent.Itmightbenoted
            that many other business fields have conducted the same type of analysis during their
            evolution. For example, Information Systems has had a long history of attempting to
            define its boundaries of research (Agarwal and Lucas, 2005; Benbasat and Zmud, 2004;
            Benbasat and Weber, 1996; Orlikowski and Iacono, 2001; Whinston and Geng, 2004),          Operations
            even though it is an extremely young field.                                              management
                                                                                                         research
            Journal selection
            We wished to expand the journals included in the study beyond the three in the
            Meredith et al. (1989) paper. We used Barman et al. (2001) which captured journal                 713
            perceptions from US scholars and Soteriou et al. (1999) which captured European
            perceptions. Specifically, we took the top journals from Table 5 in both Barman et al.
            (2001, p. 376) and Soteriou et al. (1999, p. 232). We felt that by including journals that
            are highly regarded by both US and European scholars, our research would be less
            biased toward any single view of acceptable OM research outlets. The final list of
            journals included the original three – MS, DS, and JOM – as well as two other
            well-recognized journals: IJOPM, and Production and Operations Management (POM).
            It should be noted that by selecting highly-regarded (i.e. by established scholars)
            outlets, we are potentially biasing our results against new approaches (i.e. the choice
            and use of various research methods). Hence, our results about the evolution of OM
            research could fall on the more conservative side of change.
            Article inclusion
            Owingtothe journals selected (i.e. DS and MS are interdisciplinary), it was necessary
            to extract articles that were primarily OM rather than another discipline. Our approach
            was to start with the definitions of OM research employed by other researchers
            (Young et al., 1996; Malhotra and Kher, 1996) as our initial set of criteria for article
            inclusion. The initial criteria involved lengthy discussion and underwent several
            revisions based on the results of several pilot rounds.
               We first decided that each article had to contribute to only OM research, thus
            eliminating introductions to special issues, articles that focused on teaching or
            curriculum issues (including research on teaching), and articles that focused on the
            discipline of the OM field itself such as publication productivity rankings and journal
            rankings. To be included in our study, an article’s focus must either be on an OM topic
            (see A, below) or on OM research itself (B):
               A. Similar to Malhotra and Kher (1996), the article’s major emphasis had to fall within the
               OperationsManagementBodyofKnowledge(OMBOK),asdefined(e.g.Youngetal.,1996)by
               the major topics in OM textbooks, rather than within management science, engineering,
               economics, or other allied field. “Major emphasis” means the primary focus of the paper,
               rather than the strict number of pages, title, intent of the work, types of references, or some
               other mechanical characteristic of the paper. In those often difficult cases where a
               quantitative model was a substantial portion of the paper, the article would be considered an
               Operations Management paper if either substantial insights were provided into the OMBOK
               in terms of better understanding the relationships and concepts that form the foundations of
               Operations Management, or substantial guidance was provided for OM managers based on
               the outcome of the research.
               B. The article had to analyze and contribute to the research being conducted in OM, though
               not necessarily restricted to a specific topic in the OMBOK. This would thus include analyses
               of the literature and research methodologies as long as the analysis culminated in substantial
               recommendations for improving OM research.