Trends and challenges in operations strategy research: Findings from a systematic literature review

This article is distributed under the terms of the Creative Commons Atribution 4.0 Internacional License institutua enpresa Instituto de Economía Aplicada a la Empresa Management Letters / Cuadernos de Gestión journal homepage: http://www.ehu.eus/cuadernosdegestion/revista/es/ ISSN: 1131-6837 / e-ISSN: 1988-2157 Management Letters Cuadernos de Gestión Enpresa Institutua, UPV/EHU Conocimiento en Gestión/Management Knowledge Volume 22 / Number 2 (2022) • ISSN: 1131-6837 / e-ISSN: 1988-2157 http://www.ehu.eus/cuadernosdegestion/revista/es/ Trends and challenges in operations strategy research: Findings from a systematic literature review Tendencias y retos de investigación en estrategia de operaciones: hallazgos de una revisión sistemática de la literatura Jorge A. Vivares*, Lucía Avella a , William Sarache b a Departamento de Administración de Empresas, Universidad de Oviedo. Avda. del Cristo s/n, 33006. Oviedo, España–[email protected]–https://orcid.org/0000-00032598-7318 b Universidad Nacional de Colombia, Sede Manizales, Facultad de Ingeniería y Arquitectura, Departamento de Ingeniería Industrial. Cra. 27 # 64-60, Campus Palogrande, 170004. Manizales, Colombia–[email protected]–https://orcid.org/0000-0003-3543-4151 * Corresponding author: Departamento de Ingeniería Industrial, Universidad Nacional Abierta y a Distancia. Diagonal 25F - Cra 23, 661001. Dosquebradas, Colombia–[email protected]–https://orcid.org/0000-0002-4818-5249 ARTICLE INFO Received 28 June 2021, Accepted 05 January 2022 Available online 4 March 2022 DOI: 10.5295/cdg.211543ja JEL: L21, M11 ABSTRACT Operations strategy is a mature field and requires novel insights for further research and practitioners in companies. To fill this gap, 280 articles were analyzed, by way of a systematic literature review, considering two approaches: topic perspectives (content, process, and competitive priorities), and research paradigms (analysis perspective, general methodology, research design, research typology, and alternative paradigms). Eight challenges were identified and discussed to make novel contributions to cutting-edge knowledge. Among other things, it was found that people’s perceptions of object reality prevail as the dominant paradigm, regarding the source and type of information used. In this sense, the literature suggests additional investigation through the direct observation of object reality and experimental studies, in which the action research approach could play an important role as an alternative paradigm. Additionally, considering Industry 4.0 advances, new research opportunities have emerged which permit artificial reconstruction of object reality to support decision-making. The main contribution of this study is to discuss eight challenges by which to improve research relevance and make both academic and practical contributions. Moreover, several useful data for researchers are provided, including statistics regarding response rates in survey research (dominant paradigm). Finally, these findings can be used to perform further research with additional functional strategies in the companies. Keywords: Operations Strategy, Manufacturing Strategy, Research Paradigms, Systematic Literature Review, Action Research. Management Letters / Cuadernos de Gestión 22/2 (2022) 81-96 82 Jorge A. Vivares, Lucía Avella, William Sarache RESUMEN La estrategia de operaciones es un campo de estudio maduro, pero requiere perspectivas novedosas para la investigación futura y para los profesionales en las empresas. Para llenar este vacío, se analizaron 280 artículos mediante una revisión sistemática de la literatura, bajo dos enfoques: tópicos (contenido, proceso, prioridades competitivas) y paradigmas de investigación (perspectivas, metodología, diseños, tipologías y paradigmas alternativos de la investigación). Se identificaron y discutieron ocho retos para hacer contribuciones novedosas en las fronteras del conocimiento. Entre otras cuestiones, respecto a la fuente y tipo de información utilizada, se encontró que el paradigma dominante de investigación se relaciona con las percepciones de la gente sobre la realidad objeto de estudio. En este sentido, la literatura sugiere incrementar la investigación a través de la observación directa de la realidad y los estudios experimentales, en lo cual la investigación acción puede jugar un rol importante como paradigma alternativo. Adicionalmente, considerando los avances de la Industria 4.0, han emergido nuevas oportunidades de investigación que permiten la reconstrucción artificial de la realidad para apoyar la toma de decisiones. La principal contribución de este estudio es la discusión de ocho retos para mejorar la relevancia de la investigación y hacer contribuciones tanto académicas como prácticas. Además, se proporcionan varios datos útiles para los investigadores como la tasa de respuesta en la investigación con encuestas (paradigma dominante), entre otros. Finalmente, los hallazgos pueden ser utilizados para investigar otras estrategias funcionales para las empresas. Palabras clave: Estrategia de Operaciones, Estrategia de Manufactura, Paradigmas de Investigación, Revisión Sistemática de la Literatura, Investigación Acción. Management Letters / Cuadernos de Gestión 22/2 (2022) 81-96 Trends and challenges in operations strategy research: Findings from a systematic literature review 83 1. INTRODUCTION Corporate and competitive strategies are a starting point for functional strategy development in areas such as finances, marketing, operations, and purchasing, among others. Operations Strategy (OS) is a functional strategy “…concerned with how the competitive environment is changing and what the operation has to do in order to meet current and future challenges (Slack and Lewis 2011, p.7)”. OS is a long-term plan, in which actions to be performed are established, so as to achieve the competitive advantages derived from a manufacturing/operations system (Arana-Solares etal. 2019). OS has drawn the attention of investigators and practitioners in countries the world over, which is reflected in the high number of publications present in specialized journals and books. Owing to this growth, OS study constitutes a body of knowledge which merits permanent monitoring. As shown in Table1, four OS literature reviews have been published in the past. These addressed themes related to OS, content/process, and certain methodological issues. The analysis of the present study revealed that two relevant topics require indepth OS review because they were not encompassed in previous studies: a)trade-off and sand cone themes, and b)research paradigms. Furthermore, it is necessary to analyze research published in the past several years because Chatha and Butt (2015) and Chatha etal.’s (2015) studies included articles published in and prior to 2010 exclusively, and only considered “manufacturing strategy” key in their main strategy search (herein, however, the evolution from manufacturing strategy to operations strategy is discussed, and both terms were used in the strategy search). Table 1 Previous OS literature review studies Analysis criteria Authors Dangayach and Deshmukh (2001)Boyer etal. (2005)Chatha and Butt (2015)Chatha etal. (2015)This study Source of the search 31 journals and international conferences Focused on one journal (POM) Business Source Premier. References in four articles Business Source Premier Scopus and Tree of Science (ToS) “Manufacturing strategy” term * * * * * “Operations strategy” term * * * Number of articles analyzed 260 31 506 512 280 Annual rate of articles analyzed 8.1 2.4 11.2 11.4 16.5 Number of journals encompassed 21 journals and 10 international conferences 1 34 34 105 Counts: By topic Content * * * * Process * * * * Content and process * * * Trade-off and sand cone models * By research paradigm * General methodological approach(1) * * * * * Specific research design(2) * * Specific method or technique * * By research typology(4) * Research alternatives paradigms(3) * Notes: (1) Qualitative, quantitative, or mixed. (2) Not experimental, experimental. (3) Based on Meredith etal. (1989) and Craighead and Meredith’s (2008): natural/artificial and rational/existential dimensions. (4) Two typologies: classical, and Phillips and Pugh’s (2010) improved typology. *Indicates element presence. Source: Author elaboration based on the cited references. Management Letters / Cuadernos de Gestión 22/2 (2022) 81-96 84 Jorge A. Vivares, Lucía Avella, William Sarache The present article expounds upon the results of a systematic literature review, which attempts to close the above-mentioned gaps. The underpinning of the search strategy was based on queries made in the Scopus database, and the use of a tool called Tree of Science (ToS). A total of 280 articles were analyzed. As a main contribution, eight challenges were identified and discussed for further research, in order to make novel contributions to cutting-edge knowledge. For example, on analysis of research paradigms, as proposed by Meredith etal. (1989), it was found that the current dominant OS paradigm is “People’s perceptions of object reality,” as well as the “Logical positivist/empiricist” paradigm. Findings suggest increasing the presence of alternative research paradigms (but not the replacement of the current dominant paradigm), based on “direct observation of object reality,” and “artificial reconstructions of reality”. The remainder of the article is structured as follows: in Section2, a conceptualization of the OS field, as well as the corresponding paradigms, are presented to lay a theoretical foundation for the review. The investigation’s methodology is explicated in Section3. In Section4, the main results identified are exhibited. Finally, discussion and conclusions are presented in sections5 and 6, respectively. 2. BACKGROUND 2.1. From Manufacturing Strategy (MS) to Operations Strategy (OS) A number of authors recognize that MS emerged as a field of study following the seminal contributions of Wickham Skinner (1966, 1969). Subsequently, further relevant investigations, published by Skinner (1974), Hayes and Schmenner (1978), Wheelwright (1978), Schmenner (1982), Hayes and Wheelwright (1984), and Wheelwright (1984), among others, aided in the consolidation of the MS’ theoretical foundations. Despite its status as a mature field, Kulkarni etal. (2019) assessed the paradox in the difference between academic and industry definitions, using text mining, and found some misalignment. The term “manufacturing strategy” was originally used, due to its link with companies that produced goods. For Wheelwright “…a manufacturing strategy consists of a sequence of decisions that will enable a business unit to achieve its desired competitive advantage (Wheelwright 1984, p.85)”. However, as the service sector became stronger, this concept evolved, until it became “Operations Strategy”. This concept entails an action field in the production of goods (tangible) and services (intangible), in all types of organizations (service and manufacturing companies, large and small, forand not-for-profit organizations). According to Slack and Lewis (2011, p.7) “Operations strategy is concerned less with individual processes and more with the total transformation process that is the whole business” and this implies similarities at a strategic level in different types of companies. Two components should be addressed in OS study: content and process (Leong etal. 1990). OS content defines its performance objectives, as well as the form in which operations system strategic decision areas are to be adjusted (Drohomeretski etal. 2014). The two fundamental elements of OS content study are: competitive priorities and strategic decision areas. Competitive priorities refer to an operations system’s performance objectives. In the literature, the most common competitive priorities are: cost, quality, flexibility, innovativeness (product), delivery, service, and environmental protection (da Silveira and Sousa 2010; Vivares etal. 2019). In order to manage competitive priorities, there are two dominant models to be analyzed: the trade-off (Skinner 1969, 1974), and the sand cone (Ferdows and De Meyer 1990). The trade-off model proposes the need to emphasize one or few competitive priorities over others, as it is difficult to achieve strong performance in them all (Da Silveira and Slack 2001). The sand cone model proposes a capability accumulation strategy, which, from a continuous improvement approach, permits the achievement of strong performance in all competitive priorities, maintaining the quality→delivery→flexibility→cost sequence. Strategic decision areas represent subsystems which must be modified in order to develop distinctive capacities. These lead to compliance with performance objectives (competitive priorities). Strategic decision areas in the operations system include human resources, production planning and control, sourcing, process technology, facilities, and products (Miltenburg 2008). In the OS process, the way in which OS is formulated and implemented is analyzed. A minimum of two components must be considered: 1) the pattern to direct decision-making, and 2) the model, methodology, or procedure for OS formulation and implementation. There are three main approaches in the literature for decision-making directions: top-down, bottom-up, and middle-out (Kim etal. 2014). In the OS formulation process, social, political, cultural, and learning values all come into play. Also, stakeholders’ values, needs, and interests must be considered, which makes this a highly complex process. Its complexity indicates the need for the adoption of a model (Miltenburg 2008), methodology (Hill 2000), or procedure (Dangayach and Deshmukh 2001) adequate for OS design and development. Furthermore, some variables may be simultaneously associated with OS content and process, as was found by Vivares etal. (2019, p.340) respect to Manufacturing’s Strategic Role (MSR): “...is a missing link variable in the diverse OS frameworks studied… a renewed idea of OS is proposed in this study, which adds MSR as a fundamental element to the traditional content and process-based view”. 2.2. Research paradigms To begin, the concept of OS paradigms will be discussed, so as to define their use within the present document. The concept of paradigms is quite broad and has long merited discussion. Though the term was used in ancient Greek, Thomas Khun encouraged it in his celebrated essay entitled “The structure of scientific revolutions” and provided several definitions. Based on Corbetta’s (2003) contribution, a paradigm may be understood as a theoretical perspective which has been accepted by the scientific community, is founded on previous knowledge from a given discipline, directs research through Management Letters / Cuadernos de Gestión 22/2 (2022) 81-96 Trends and challenges in operations strategy research: Findings from a systematic literature review 85 the specification of what to study, formulates hypotheses about a phenomenon, and selects the most suitable research techniques. Thus, paradigms are well-established beliefs about theories, methods, or approaches, which are phenomena or epistemological ways to study a given phenomenon and acquire knowledge. There are several views from which to analyze paradigms in the OS field. For example, Voss (1995) reviewed the history of manufacturing strategy, and discussed theoretical paradigms which included key success factors, order winners, capabilities, process choices, contingent approaches, infrastructure approaches, and world class manufacturing, among others. From the epistemological point of view, Westbrook (1995) advocated that action research was a new paradigm for research in production and operations management, and that it constituted more than a single technique or method. In this study, special attention was paid to the framework for paradigm comprehension, as proposed by Meredith etal. (1989), in an article entitled “Alternative Research Paradigms in Operations”. Meredith etal. (1989) developed a framework for paradigms of research methods with two dimensions: the rational/existential and the natural/artificial. The first “relates to the epistemological structure of the research process itself (Meredith etal. 1989, p.305)” and refers to three ways that a researcher may observe reality: direct observation of object reality, people’s perceptions of object reality, and artificial reconstruction of object reality. The second dimension “…concerns the source and kind of information used in the research (p.307)” and refers to four epistemological ways to analyze observations: from the axiomatic, logical positivist/empiricist, interpretative, and critical theory. Despite its age, Meredith etal.’s (1989) framework remains relevant for the novel study of paradigms. The aforementioned authors used their framework to review the operations management state of the art, by employing a sample of articles published in a span of just two years (1977 and 1987), when Operations Strategy (OS) was just gaining traction as a new field of study. They observed that Operations Management (OM) investigation had concentrated on the “axiomatic” category of the rational/existential dimension, and the “artificial construction of reality” in the natural/artificial dimensions. Thereafter, Craighead and Meredith (2008) used the same framework to analyze a sample of OM articles from 1995 and 2003 (separately). They found little movement in people’s perceptions of the object reality category in the natural/artificial dimensions, and to the “logical positivist/empiricist” category in the rational/existential dimensions. However, these studies addressed the OM field, and no studies were found to have used this framework for in-depth paradigm reviews in the OS field. 3. METHODOLOGY Principles for the performance of a systematic literature review were applied (Bartels 2013; Granillo and González 2021; Orviz et al. 2021). Figure 1 shows the general methodology. First, the target of the review was delimited to answer two main questions: 1)What are the general features of the articles published on OS content and process (themes, methodologies, paradigms)? 2)What are the main areas in which further research may be conducted? Delimit target of the review Create and refine a search strategy Read and classify the articles found Is the search strategy suitable? Yes No Analyze and present the results Apply the search strategy Figure 1 General methodology used in the systematic literature review Source: Author elaboration. Second, a search strategy was created and refined in an iterative process, in accordance with two criteria (Bartels 2013): 1)relevance of the articles found, and 2)balance between quantity (sensitivity) and specificity. For this purpose, 22relevant references were used. The refinement process was halted only while the search strategy identified these references. Third, the final search strategy was applied. The strategy had two main components: on the one hand, a search equation was created (Eq. 1) and applied in the Scopus database. This expression may be considered an equation which results in a C set with n references. These depend on source (database and search fields), the keywords, the time, as well as the document type (C={r1, r2, ..., rn}=f(source, keywords, time, document type)). The equation was applied using the TITLE-ABS-KEY field (title, abstract and keywords), beginning with the year 2001, and was limited by document type (LIMIT-TO (DOCTYPE)), article (ar), or review (re). Boolean operators (OR, AND) were used to join the keywords. f(source, keywords, time, document type)=TITLEABS-KEY (“manufacturing strategy” OR “operations strategy”) AND TITLE-ABS-KEY (“competitive priorities” OR “decision areas” OR “decision categories” OR levers OR taxonomy OR (tradeoff) OR “sand cone” OR (formulation process*) OR (implementation process*) OR (formalization process*) OR (formation process*) OR (choice* process*) OR (choice* implementation) OR (strategic choice*) OR “stage model” ) AND TITLE-ABS-KEY (survey OR “case study*” OR (literature review) OR framework OR interview* OR “conceptual model” OR “conceptual discussion” OR “action research” OR empirical) AND PUBYEAR ≥ 2001 AND (LIMITTO (DOCTYPE, “ar”) OR LIMIT-TO (DOCTYPE, “re”)). (1) Management Letters / Cuadernos de Gestión 22/2 (2022) 81-96 86 Jorge A. Vivares, Lucía Avella, William Sarache On the other hand, the Tree of Science (ToS)1 methodology, proposed by Robledo etal. (2014), was used in order to minimize bias. ToS is based on graph theory and detects relevant references, while minimizing the bias which results from searches performed in specific databases, and has been applied in recent research (Durán et al. 2021; Robledo et al. 2021). Subsequently, the search strategy was applied, and three criteria for inclusion/exclusion were established: a)the inclusion of articles related to OS management and strategic decisions, and exclusion of those related to technical issues in manufacturing or operations (e.g., physical transformation), b)the inclusion of articles, if the full text version was accessible to authors, c)the exclusion of duplicates. In order to classify the selected references, a protocol review was created, as suggested by Kitchenham (2004). The articles were read and classified by typology (theoretical or evidentiary). Theoretical articles were classified into one of three categories: systematic reviews, conceptual articles or frameworks, and essays or reflections. Evidentiary articles were classified by topic (OS content and process, trade-off and sand cone models), and by research paradigm (see Table2). In order to ensure reliability and a certain level of concordance among authors, an article classification audit was performed. Thus, doubts were discussed, a consensus was reached, and the review protocol was improved. Results from the analyses are presented below. Table 2 Research paradigms for investigation performance (evidentiary articles) Category Subcategories Analysis perspective Quantitative, qualitative, or mixed Research design Not experimental (data: cross-sectional, longitudinal, panel, qualitative), or experimental General methodology Survey/interview (sample), case studies, others (action research, artificial evidence), or multiple Research typology 1 Historical, descriptive, correlational, explanatory, exploratory, or qualitative Research typology 2 Exploratory, problem-solving, testing-out, or theory-building. The first three were proposed by Phillips and Pugh (2010), and theory building was extracted from the literature (Eisenhardt and Graebner 2007; Wacker 1998) to complement the first proposal. Research method paradigms, based on those of Meredith etal. (1989) and Craighead and Meredith (2008) 1)Natural/artificial dimension: direct observation of object reality, people’s perceptions of object reality, or artificial reconstruction of object reality. 2)Rational/existential dimension: axiomatic, logical positivist/empiricist, interpretative, or mixed. Source: Author elaboration. 1 https://tos.coreofscience.com/ 4. RESULTS A total of 319unduplicated references were detected. Of these, 39 were discarded, as they did not comply with the inclusion criteria. This left 280 articles for the final analysis, 14.3% of which were theoretical (40), and the other 85.7% of which corresponded to 240articles which incorporated evidence (whether empirical or artificial). Said 280articles had been published in 105different journals (between 2001 and 2017). Just eight of these journals had published 51% of said contributions. In a supplementary file, detailed information about the 280 articles analyzed is provided. The results revealed that an average of 16.5 OS articles are published per year (DS=1.234). 4.1. Results of theoretical contributions Table 3 shows descriptive statistics about the theoretical contributions. Articles which conceptualized a topic and/or proposed frameworks were dominant. Two essays (or reflections) appeared: one in recognition of Bob Hayes’ contributions (Fisher 2007), and another written by Skinner (2007), called “Manufacturing strategy: The story of its evolution”. Table 3 Theoretical articles Categories Results Total % Row Quantitative Qualitative Systematic review 9 2 11 27.5 Conceptualizations or frameworks  27 27 67.5 Essays or reflections  2 2 5.0 Total 9 31 40 100 % Column 22.5 77.5 100  Source: Author elaboration. 4.2. Evidentiary articles (empirical or artificial evidence) A total of 240 articles detected provided empirical or artificial evidence regarding their respective objects of study. Below, the results are presented, organized by topic addressed and research paradigm. A. Results by topic addressed From a thematic point of view, OS investigation has focused on content (74%), while just 6% has been oriented toward the study of process. The remaining 20% has simultaneously studied process and content variables. In Figure 2, the number of investigations addressing process, content, or both topics, is presented. For those studies focused on process, in addition to being the scarcest type, they were also the most intermittent, given that not a single contribution was registered in seven of the 17years analyzed. Management Letters / Cuadernos de Gestión 22/2 (2022) 81-96 Trends and challenges in operations strategy research: Findings from a systematic literature review 87 0 2 4 6 8 10 12 14 16 18 20 01 02 03 04 05 06 07 08 09 10 11 12 13 14 15 16 17 Count Year Content Content & process Process Figure 2 Longitudinal results by OS topic Source: Author elaboration. Based on 63articles including process variables, it was noted that certain contributions, which studied models, methodologies, or procedures that supported decision making, were notable (67%). However, it was established that the majority of said investigations addressed specific OS decisions, and did not consider matters regarding its general configuration, formulation, or implementation. Topics, such as pattern analysis in decision-making and diverse topics involved in the process, have received scant attention in the literature: top down (16%), bottom up (16%), and mixed (6%). 83% 83% 80% 79% 37% 27% 14% 29% 0% 20%40%60%80%100% Cost Quality Delivery Flexibility Innovativeness Service Environmental protection Others n=150 3a. Competitive priorities Trade-off 9% Sand cone 7% Neither 84% 3b. Models tested Figure 3 Competitive priorities and their analytical paradigms Source: Author elaboration. Contributions that have addressed OS content variables (n=225) have concentrated principally on the study of competitive priorities (67%), and in lesser proportions, on one or several strategic decision areas (52%). In Figure3 a summary of articles which address competitive priorities are presented. Note that cost, quality, flexibility, and delivery are the most recurrent (Figure3a). Furthermore, it was found that 9% of the articles empirically tested the trade-off model, and 7% tested the sand cone model (Figure3b). B. Research paradigm results The results presented in Table4 indicate that the majority of investigations were based on surveys/interviews applied in companies, with employees or experts (55%). In investigations that used the surveys applied in companies (n=117), the average response rate was 33.7% (minimum: 5.6%, maximum: 86.1%) and the quartile distribution is as follows: quartile 1 (17.9%), quartile2 (26.4%), quartile3 (45.3%), and quartile4 (86.1%). This result may be useful for future studies based on survey research. Case studies represented 29.6% of the total, and other types of investigation (action research, and artificial evidence, among others), together with the use of multiple approaches, represented 7.5% and 7.9%, respectively. From the perspective of analysis, studies with quantitative approaches represented 56.7% of the investigations, followed by qualitative approaches, which constituted 22.9%. Finally, 20.4% of the investigations used mixed perspectives. Following the longitudinal evaluation of these variables, no patterns or tendencies were found. In regard to research design, the majority of the studies were not experimental (98%). In this category (see Table5), investigations with transversal data were prevalent. There were few longitudinal contributions. Studies with panel data were not detected either. However, five experimental investigations were found, which, although marginal, indicate that investigation from this perspective is indeed possible. Said studies were completed in the early 2000s (Pullman etal. 2001; Stading etal. 2001; Lee etal. 2002; Li etal. 2006; AlDurgham and Barghash 2008). Research typology showed that most of the studies gathered were either descriptive (25.9%) or explanatory (38.3%) (see Table6). In fact, no study regarding OS history was found at all. From the perspective of Phillips and Pugh (2010), testing-out methods, those which attempted to test or refute theories, conformed the majority (49.2%), while the least frequent were those of the theory-building type. Management Letters / Cuadernos de Gestión 22/2 (2022) 81-96 88 Jorge A. Vivares, Lucía Avella, William Sarache Table 4 General methodology and analysis perspective Category Analysis perspective Total % Row Quantitative Qualitative Mixed Survey/interview (sample) 117 1 14 132 55.0 Sample of companies 106  12 118 49.2 Sample of employees 4 1 2 7 2.9 Sample of experts 7   7 2.9 Case studies 8 46 17 71 29.6 Single case study 3 13 11 27 11.3 Multiple case study (2-4) 4 7 2 13 5.4 Multiple case study (5 or more) 1 26 4 31 12.9 Others 4 6 8 18 7.5 Action research  6 6 12 5.0 Artificial evidence 3  2 5 2.1 Others 1   1 0.4 Multiple 7 2 10 19 7.9 Sample of companies and employees 7  1 8 5.0 Sample of companies and case study   7 7 2.9 Sample of companies and experts   1  0.0 Sample of experts and case study  2  2 0.8 Sample of experts and employees   1 1 0.4 Total 136 55 49 240 100 % Column 56.7 22.9 20.4 100  Source: Author elaboration. Table 5 Research design Category Total % Data gathering methods (counts) (*) A B C D E F Not experimental 235 98 160 93 31 22 20 43 Cross-sectional 169 70 145 44 12 15 17 20 Longitudinal 11 5 8 4   3 1 Panel 0 0       Qualitative 55 23 7 45 19 7  22 Experimental 5 2 3 2 0 1 0 2 Total 240 100 163 95 31 23 20 45 Note: (*) A (surveys), B (interviews), C (documentary analyses), D(panels or focus groups), E(existing databases), F(field observations/ measurements). Source: Author elaboration. Table 6 Research typology Typology 1 Typology 2 Total % row Exploratory Problemsolving Testingout Theorybuilding Historical 0 0.0 Descriptive 32 9 17 4 62 25.9 Correlational 6 2 8 1 17 7.1 Explanatory 1 3 87 1 92 38.3 Exploratory 6 5 1 2 14 5.8 Qualitative 25 10 5 15 55 22.9 Total 70 29 118 23 240 100 % column 29.1 12.1 49.2 9.6 100  Source: Author elaboration. In accordance with Meredith etal.’s (1989) approach to the examination of research paradigms, OS investigation predominantly represents positivist/empiricist logic, with an approach to reality through people’s perceptions of object reality (Table7). As observed in Figure4, from the natural/artificial dimension, those studies with direct observation of object reality doubled in the 2009-2017 period, as compared to previous years. Tendencies were not found in the rational/existential dimension, and it should be emphasized that the two studies from the axiomatic perspective corresponded to 2002 and 2016 (Choy etal. 2016; Lee etal. 2002). Management Letters / Cuadernos de Gestión 22/2 (2022) 81-96 Trends and challenges in operations strategy research: Findings from a systematic literature review 89 Table 7 OS research method paradigms Dimensions Natural/Artificial Total % Row Direct observation of object reality People’s perceptions of object reality Artificial reconstruction of object reality Rational/ Existential Axiomatic   2 2 0.8 Logical positivist/empiricist 4133 11 148 61.7 Interpretative 22 46 5 73 30.4 Mixed (positivist + interpretative) 1 13  14 5.8 Mixed (axiomatic + interpretative) 2  1 3 1.3 Total 29 192 19 240 100 % Column 12.1 80.0 7.9 100  Source: Author elaboration. 01 02 03 04 05 06 07 08 09 10 11 12 13 14 15 16 17 AR 0 10 6 6 14 11 24 21 9 0 5 0 13 0 0 7 7 PP 83 80 82 82 86 89 58 79 64 91 81 94 81 70 64 79 93 DO 17 10 12 12 0 0 18 0 27 9 14 6 6 30 36 14 0 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% Year Figure 4 General results by paradigm (natural/artificial dimensions) Source: Author elaboration. Note: AR (artificial reconstruction of object reality), PP (people’s perceptions of object reality), DO (direct observation of object reality). The specific methods used by the encountered investigations, in order to obtain results, have been predominantly statistical. Figure5 shows that the most used techniques included: differences in means, regression analysis, Structural Equation Modelling (SEM), cluster analysis and measures of association. Also, 12articles were found to have designed or applied aggregate performance indicators. Diverse techniques and other mathematical theories have been little used in OS study (e.g., fuzzy logic, optimization, network analysis, etc.). 22,1% 17,1% 12,5% 11,7% 9,6% 6,3% 4,6% 3,3% 3,3% 2,5% 2,5% 1,7% 1,3% 0,8% 0,8% 0,4% 0,0% 0,0% 17,1% 0% 5% 10%15%20%25%30% Difference in means Regression analysis SEM Cluster analysis Measures of association Expert evaluation Grounded theory Canonical correlation AHP/ANP Fuzzy logic Mathematical model Simulation Ethnography Optimization QFD Network analysis Laboratory experiment Delphi method Others n=240 Figure 5 Methods used Source: Author elaboration. 5. DISCUSSION AND FUTURE RESEARCH AREAS OS investigation has been relatively prolific during the past few years. However, it is a field with great potential and multiple possibilities for future investigation. When considering the multiple, complex phenomena which intervene in a company’s production/operations system, the possibilities for continued OS investigation appear to be unlimited. Below, a discussion will attempt to identify future areas, in the following order: topic perspectives and investigative paradigms for further research. 5.1. Topic perspectives OS investigation continues to concentrate principally on content, and very little on the formulation process. This demonstrates the continuity of the tendency encountered by Dangayach and Deshmukh (2001). The OS formulation process is fertile ground for investigation, as there is no universal strategy which can be employed by all. Thus, each company must establish their own path to excellence (Schroeder and Flynn 2001). Miltenburg (2009, p.6180) affirms this, stating that: “Many models are possible and there is no particular model that is best for all companies”. Further investigation into the OS formulation process is required, particularly regarding matters such as: a)design and testing of models/methodologies/procedures for strategy formulation and implementation, b)the study of patterns in decision-making (top-down, bottom-up, and mixed), c)the evaluation and comprehension of different variables which affect the formulation process, such as: external, cultural, political, financial, stakeholders, learning, and discussion-interaction mechanism factors, in addition to project management for improvement of operations systems, among other things. Competitive priorities have been studied chiefly as operations system performance indicators. Very few investigations, however, have simultaneously evaluated the degree of importance which each represent for company strategy. It is necessary to separate the capacities that a company wishes to strengthen in the future (priorities) in accordance with their importance for strategy, and their current ability/strength/capability to compete in the market (Corbett and Claridge 2002; Nauhria etal. 2011; Management Letters / Cuadernos de Gestión 22/2 (2022) 81-96 96 Jorge A. Vivares, Lucía Avella, William Sarache Xu, X., 2012. From cloud computing to cloud manufacturing. Robotics and Computer-Integrated Manufacturing, 28(1), 75-86. DOI: 10.1016/j.rcim.2011.07.002 Yu W., Chavez R., Feng M., Wong C.Y. and Fynes B., 2020. Green human resource management and environmental cooperation: An ability-motivation-opportunity and contingency perspective. International Journal of Production Economics, 219, 224-235. DOI: 10.1016/j.ijpe.2019.06.013 Zhou D., Yan T., Zhao L. and Guo J., 2020. Performance implications of servitization: Does a Manufacturer’s service supply network matter? International Journal of Production Economics, 219, 31-42. DOI: 10.1016/j.ijpe.2019.05.019 Management Letters / Cuadernos de Gestión 22/2 (2022) 81-96