The impact of lean manufacturing on innovation performance: The moderation effect of product modularity and internal integration

The impact of lean manufacturing on innovation

performance: The moderation effect of product modularity

and internal integration

Master Thesis

MSc Supply Chain Management

University of Groningen, Faculty of Business and Economics

June 22

, 2018

TOM BRÜGGENBROCK

Studentnumber: 3511049

e-mail: t.brueggenbrock@student.rug.nl

Supervisor / University of Groningen

Dr. Ir. Thomas Bortolotti

Co-assessor / University of Groningen

Dr. Kristian Peters

Acknowledgments: I would like to thank Dr. Ir. Thomas Bortolotti for his time, guidance and

The impact of lean manufacturing on innovation

performance: The moderation effect of product modularity and

internal integration

A B S T R A C T

Embedded in an increasing competitive intensity due to globalization, the multifaceted construct of innovation gains attention not only on a strategic but also on an operational level; also in rela-tion to lean manufacturing. Nonetheless, academia missed out to fully understand the relarela-tionship dynamics between all facets of lean manufacturing and innovation. In particular, the link of Just-In-Time (JIT) and innovation performance requires further research. Thus, this thesis provides an empirical research which analyses the relations between Total Quality Management (TQM), JIT, and Infrastructure practices and innovation. It seeks not only to contribute knowledge about the main effects of lean on the performance construct but also to evaluate the impact of two contextual factors on any lean-innovation relationship; namely internal integration and product modularity. Based on data of 317 plants accessed through the High Performance Manufacturing (HPM) pro-ject, the analysis was performed by Structural Equation Modeling (SEM) and revealed two major findings. First, infrastructure practices encourage innovation significantly. Second, internal inte-gration weakens the positive impact of infrastructure practices, but enhances the TQM-innovation relationship.

Keywords: lean; total quality management; just-in-time; infrastructure; innovation performance; product

INTRODUCTION

International supply chain structures and complex market dynamics force organizations to rethink and adapt manufacturing strategies to stay competitive (Simchi-Levi, Peruvankal & Mulani, 2012). Organizations are pushed to respond quickly to shifts in customer demand, to insist on the highest quality standards and to cut cost to the extreme. But, is it viable for organizations to gain a competitive advantage in all above aspects?

With Nakane (1986) and Ferdows and De Meyer (1990) a research stream emerged in which organizations are viewed as being able to increase their capabilities in terms of quality, delivery, flexibility and cost in a cumulative manner (Bortolotti, Danese, Flynn & Romano, 2015; Roger G. Schroeder, Shah & Xiaosong Peng, 2011; Brown, Squire & Blackmon, 2007; Flynn & Flynn, 2004). However, prior literature discusses likewise an approach suggesting the polar opposite: the trade-off model (Skinner, 1969). It acknowledges the conflict of objectives. Organizations are hold to prioritize and develop within single capabilities that suit their strategic objectives best (Boyer & Lewis, 2002; Safizadeh, Ritzman, Sharma & Wood, 1996). Thus, similar to the allocation of resources (Venkatraman & Prescott, 1990) and determination of business strategies (Porter, 1996), organizations need to make a choice and trade off among capabilities; especially, when their operations lie on the efficient frontier (Hopp, 2011; Roger G. Schroeder et al., 2011). In that response, this thesis considers conflicts in an organizations’ capability to excel in the areas of quality, cost and innovation simultaneously. Although, a certain amount of research provides empirical evidence about the role of quality as basis for innovation (Zeng, Phan & Matsui, 2015; Prajogo & Sohal, 2003) and cost efficiency (Narasimhan & Schoenherr, 2013; Flynn * & Flynn, 2005), organizations still have to trade off quality and innovation against cost savings at a certain stage (Safizadeh, Ritzman & Mallick, 2000; Noble, 1995). On the one hand, recognizable in form of prevention or appraisal activities, additional costs occur by improving quality at a certain point (Farooq, Kirchain, Novoa & Araujo, 2017). On the other hand, short product life cycles require organizations to overlap development initiatives and to ignore cost benefits of sequential product development, e.g. finalizing design activities first before specifying production characteristics further downstream the value chain (Roemer, Ahmadi & Wang, 2000). A trade-off that needs to be mastered since successful innovations define a key for competitive advantage (Pisano, 2015). Especially in everchanging environments, organizations are pushed towards innovations and depended on their positive effect on firm performance (cf. Zhou, Yim & Tse, 2005).

sense, it implies the creation of streamlined operations systems capable to produce perfect quality (Slack, Chambers & Johnston, 2010). On the operational level, the philosophy translates into so-cio-technical practices (Shah & Ward, 2003). This study focuses on the set of practices associated with Total Quality Management (TQM) and Just-in-Time (JIT) and also combines the ones asso-ciated with Total Preventive Maintenance (TPM) and Human Resource Management (HRM) into a single practice bundle called Infrastructure (INFRA) similar to certain other studies in OM lit-erature (Bortolotti, Danese, et al., 2015; Cua, McKone & Schroeder, 2001). By referring to Shah and Ward (2003), the two former bundles are interrelated and internally consistent practices. However, at the same time, (Vuppalapati, Ahire and Gupta (1995) advise a joint implementation of JIT and TQM. Through a certain degree of interaction, both are mutually supportive. For in-stance, the effectiveness of JIT schedules may improve when quality levels are met and the amount of rework is reduced (Flynn, Sakakibara & Schroeder, 1995). Also, both bundles share the support of practices of the infrastructure bundle (Bortolotti, Danese, et al., 2015; Flynn, Sakakibara, et al., 1995).

Whereas operational performance has been widely studied in relation to TQM and JIT (Furlan, Vinelli & Dal Pont, 2011; Shah & Ward, 2003; Flynn, Schroeder & Sakakibara, 1995), and TQM thereto in concert with innovation (Hung, Lien, Yang, Wu & Kuo, 2011; Prajogo & Sohal, 2003), the examination of a JIT-innovation relationship is missed. Prior research fails to provide an empirical examination of JIT in relation to innovation as a single performance dimen-sion and also the joint investigation of TQM, JIT and innovation within a single framework. How-ever, it is crucial to prove what practices affect this dimension as JIT and TQM overlap and lean as an multi technique construct may form trade-off in itself. Hence, this study aims for providing a more holistic framework and provide empirical support whether lean supports innovation as a total.

Additionally, recent studies call for a context-dependent investigation of these relationships. Due to mixed results, researchers question the universal validity of JIT and TQM (Sousa & Voss, 2008) and advise to identify factors that heighten and mitigate practice-performance links (D. Zhang, Linderman & Schroeder, 2012; Mackelprang & Nair, 2010; Nair, 2006). In this sense, at the core of this research is the moderating role of product modularity and internal integration.

By conducting a hypotheses-based research, the paper contributes to the existing knowledge base on lean bundles and their likelihood to improve organizational performance. It complements the debate about the effectiveness of TQM and JIT and its shared practices and provides a more comprehensive perspective by studying both, TQM and JIT, in relation to innovation. Addition-ally, the research enriches the contingency stream in operations literature by analyzing the mod-erating effect of product modularity and integration. A theoretical contribution that further implies practical consequences. Executives gain an understanding of internal factors that mitigate lean practices and are important to be respected when adapting manufacturing and supply strategies and performance objectives. Moreover, practitioners are provided with information essential for resource allocation. Depending on the organizations strategic objectives (cost, quality, innova-tion), the paper suggests in which lean practices to invest most and which to trade off for.

LITERATURE REVIEW

Lean manufacturing

Since its beginning in the mid-seventies, the concept of JIT describes the production of “only the necessary products, at the necessary time, in the necessary quantity’’ (Sugimori, Kusunoki, Cho & Uchikawa, 1977, p. 533). At the later stage, it has been considered as both a manufacturing or managerial philosophy (cf. Upton, 1998) and as a set of practices that supports and implements the philosophy (Narasimhan, Swink & Kim, 2006; Flynn, Sakakibara, et al., 1995). Hitherto, ac-ademia does not agree on a certain batch of practices (Mackelprang & Nair, 2010). However, the concept involves socio-technical practices pursuing the total elimination of waste (Slack et al., 2010) and includes as such small lot size and set up time reduction, cellular manufacturing layout, daily schedule adherence as well as pull production (Mackelprang & Nair, 2010; Shah & Ward, 2003, 2007).

TABLE 1

Practice classification and representative literature

RESEARCH HYPOTHESES

The underlying framework of this research is presented in Figure 1. It pictures the direct effect of all three lean bundles on the performance dimension innovation. In addition, the proposed mod-eration effects of product modularity and internal integration on the lean-performance relation-ships are illustrated.

FIGURE 1

Theoretical framework

Lean - Innovation relationship

In general, research shows that the implementation of TQM leads to enhanced quality perfor-mance (Zeng et al., 2015; Prajogo & Hong, 2008; Lee, Rho & Lee, 2003; Curkovic, Vickery & Dröge, 2000), and innovation performance (Zeng, Zhang, Matsui & Zhao, 2017; Sadikoglu & Zehir, 2010; Prajogo & Hong, 2008; Prajogo & Sohal, 2003). Furthermore, the adoption of TQM leads ultimately to the reduction of rework and internal variability within the production system and, thus, to a reduction in production costs (Sila, 2007). Hence, the following hypotheses are suggested:

The positive impact of JIT on cost performance (Danese, Romano & Bortolotti, 2012; Shah & Ward, 2003) and quality performance (Bortolotti, Danese, et al., 2015; Flynn, Sakakibara, et al., 1995) is widely acknowledged in literature. However, as mentioned, in relation to innovation its effect is unknown. Nevertheless, JIT practices are considered to be especially relevant in or-ganizations facing predictable demand, i.e. operating in repetitive environments in which prod-ucts are standardized (Lander & Liker, 2007). In this light, Bortolotti, Danese and Romano (2013) examined the implementation of JIT practices at varying degrees of repetitiveness and reports that demand variability mitigates the positive impact of JIT on responsiveness. Hence, the following hypotheses are proposed:

Hypothesis 3. Just-In-Time negatively impacts on innovation performance

Moderation of Lean - Innovation relationship

Product modularity refers to the extent to which a product can be deconstructed into its independ-ent componindepend-ents and recombined (Schilling, 2000). Connected through de-coupled interfaces, the components, i.e. the product modules, can be replaced without automatically changing other parts of the product (Ulrich, 1995). In this way, the concept allows organizations to manufacture a great variety of products (M. Zhang, Zhao & Qi, 2014; Ulrich, 1995) and simultaneously exploit cost advantages through the standardization of modules (Mikkola & Gassmann, 2003). Hence, driven by Tu, Vonderembse, Ragu-Nathan and Ragu-Nathan (2004) and Jacobs, Droge, Vickery and Calantone (2011), it can be argued that modularity further supports the postponement of product customization and enhances an organizations ability to standardize and re-sequence pro-duction processes. Feitzinger and Lee (1997) also states that a module-based product architecture simplifies the diagnosis of quality problems.

In the context of innovation performance, Danese and Filippini (2010) empirically prove that the concept of modularity reduces the time needed for developing new products. On the one hand, a modular-based product architecture simplifies the upgrade of goods based on customer demand (Jacobs, Vickery & Droge, 2007; Danese & Romano, 2004) and, on the other hand, reduces the time needed for the design and test of a certain product (Araujo, Dubois & Gadde, 1999). In addition, modularity improves the innovativeness of products up to a certain level (Lau, Yam & Tang, 2011) and shows a positive impact on quality and cost (Jacobs et al., 2007). A standardized product architecture reduces investment costs, among others, in engineering and testing and re-sults in a learning curve due to a production of similar components (Fisher, Ramdas & Ulrich, 1999). Thus, the following hypotheses are formulated:

Hypothesis 8. Product modularity positively moderates the relationship between Infrastructure and innovation performance

Hypothesis 9. Product modularity positively moderates the relationship between Just-In-Time and innovation performance

The objective of internal integration is to do away with traditional silo functions by cooperation and interaction (Flynn & Flynn, 1999), thus creating better relationships between and among em-ployees. Two or more departments function as a unified entity, although they also have separate functions and aims(J. H. Y. Yeung, Selen, Zhang & Huo, 2009). For integration to happen, de-partments must share information flows and negotiate; mutual interactions and efforts are neces-sary. Hence, this research states the following hypothesis

Hypothesis 4. Internal integration positively moderates the relationship between Total Quality Management and innovation performance

Hypothesis 5. Internal integration positively moderates the relationship between Infrastructure and innovation performance

METHODOLOGY

Research database and sample

To assess the theoretical framework, I conducted a confirmatory survey research since it fits with the aim of this research best. It allows to provide empirical evidence about relations between well-defined variables and about the operational environment these are present. Also, it facilitation of findings (Karlsson, 2016). The method itself is appropriate as the knowledge base of lean manu-facturing is mature and, hence, allows academics to validate further relations among concepts and constructs and to understand structures that facilitate or mitigate the relations (Karlsson, 2016).

In line with this purpose of theory testing, this research draws on survey data gathered during the third round of the High Performance Manufacturing (HPM) project (R. G. Schroeder & Flynn, 2001). The dataset offers a basis for investigating the relationship between managerial practices and the performance of traditional and high-performance manufacturers; operating in either of the following three industries: electronics, automotive suppliers or machinery (SIC codes: 36, 37, and 35, respectively). Schroeder and Flynn (2001) emphasized that each sector was chosen with regard to its increased potential of uncertainty in plant performance and practices due to the tran-sition these sectors went through at the turn of the century. Hence, the HPM database defines an appropriate sample for the present, context-dependent research and allows me to argue for a man-ufacturing plant as unit of analysis. It is reasonable since production units traditionally typify the environment within organizations in which lean practices are used to enhance performance.

Survey questionnaire and respondent profile

To guarantee the validity of the project, the questionnaire was reviewed by experts and managers at the early stage (Bortolotti, Boscari & Danese, 2015) and translated by a native speaker in each country first and then back-translated by a different project member. Whereas the former fulfilled the purpose of content validity, the latter ensured that translations are conformed with the original English version of the questionnaires (Bozarth, Warsing, Flynn & Flynn, 2009). Additionally, to reduce common method bias, the battery of questionnaires were designed to contain reverse scales and a combination of objective and subjective measurements (Bortolotti, Danese, et al., 2015).

The final set of questionnaires were directed to a diversified group of personnel, which en-sured that most items could be answered by multiple respondents and all by the best-informed person. This procedure further minimized the key informant bias (Liu, Shah & Schroeder, 2006; Sakakibara et al., 1997) since people from various functions and positions were considered. The exact informant structure at each plant can be found in Table 3. If a question was answered by multiple of these informants, the mean of all responses was considered to perform the analysis on a plant-level as mentioned earlier (Bortolotti, Danese, et al., 2015).

Ethical principles

TABLE 2

Sample structure and statistics

Country Industry Total

Electronics Automotive suppliers Machinery

Austria 10 4 7 21 Italy 10 7 10 27 Spain 9 10 9 28 Sweden 7 7 19 24 Germany 9 19 13 41 Japan 10 13 12 35 Finland 14 10 6 30 South Korea 10 11 10 31 China 21 14 16 51 USA 9 9 11 29 Total 109 104 104 317 TABLE 3

Personnel targets of questionnaires

Respondent Number of respondent per plant

Plant accounting manager 1

Direct labor 10

Human resource manager 1

Information systems manager 1

Production control manager 1

Inventory manager 1

Member of product development team 1

Measurement scales and content validity

All scales were measured by Likert-scale questions. However, whereas the performance construct was measured in comparison to a plants’ competitors, the remaining constructs neglect the environment and follow an internal perspective – viz. items which measured performance aspects, asked informants to indicate whether the plant performs better or worse compared to its competitors on a 5-point Likert-scale (1 = poor or low; 5 = superior). Although it might be argued to use subjective measures for performance, prior research suggests the success of this method (Bozarth et al. 2009; Flynn, Schroeder, & Sakakibara, 1995; Narasimhan & Das, 2001). More-over, it is reasonable to follow this method since the sample involves various sectors and coun-tries. Perceptual items mitigate measure issues caused by different industry characteristics and enhance the comparability of plants across sectors (Bortolotti, Danese, et al., 2015; Danese, 2013).

All items that were taken into account for measuring JIT, TQM, Infrastructure and both mod-erators under study followed likewise the Likert-scale design as mentioned but requested respond-ents to indicate to what extent they agree to the statemrespond-ents on a scale from 1 (strongly disagree) to 7 (strongly agree).

As indicated, the items used in this study are a subset of the HPM survey. Based on the constructed framework in section 3, I adopted parts of the questionnaire and developed measure-ment scales from literature to ensure content validity (Karlsson, 2016). The three bundles of lean practices JIT, TQM, and Infrastructure were taken into account as second-order factors in the later stage of this research and were therefore measured by their first-order factors, i.e., the practices, which in turn were measured by a multi-item scale. In contrast, the two moderators and the per-formance construct were conceptualized as first-order latent variables and directly measured by multi-item scales.

Total Quality Management. Further, the measurement of TQM is driven by Ahmad et al. (2003) and several other studies that are shown in section 2. I adopted the TQM conceptualization of Ahmad et al. (2003) and so included statistical process control, customer involvement, supplier quality involvement and process feedback as first-order factors. However, feedback is partly considered as infrastructure practice in academia authors, i.e. as a general practice that provides a basis for practices addressing a specific performance aspect, I measured it about quality in harmony with Bortolotti, Danese, et al. (2015).

Infrastructure. With a focus on the joint practices between JIT and TQM, infrastructure is measured by five first-order factors that cover total preventive maintenance (Bortolotti, Danese, et al., 2015; Netland et al., 2015; Cua et al., 2001; McKone, Schroeder & Cua, 2001), management commitment (Cua et al., 2001) and three of the commonly shared HRM practices as shown in Table 1. Namely multi-functional employees, small group problem solving and employee sugges-tions (Ahmad & Schroeder, 2003; Cua et al., 2001; Flynn, Sakakibara, et al., 1995).

Innovation Performance. Driven by Zeng et al. (2015) this study associates innovation with the introduction of new products which are meant to meet particular customer needs (Devaraj, Hollingworth & Schroeder, 2001). In this sense, the performance construct is measured by two items validated by Zeng et al. (2015). Hence, in line with other operations management literature, it is operationalized as the speed of new product introduction (Chowdhury, Jayaram & Prajogo, 2017; Zeng et al., 2017, 2015; Terziovski & Guerrero, 2014; Noble, 1995) and product innovativeness (Zeng et al., 2015, 2017).

Product modularity. In regard to the first moderator under study, I adapted the scale of Forza et al. (2000) which has been validated and is commonly used in operations management literature (Sandrin, Trentin & Forza, 2018; M. Zhang et al., 2014; X. D. Peng, Liu & Heim, 2011). Constructed as multi-item measurement, the scale considers the commonality of parts, the appli-cation of product platforms as well as the independence of product modules.

Internal integration. The second moderator was measured by a multi-item scale validated by Heim and Peng (2010). It captures how well functions work interactively together, coordinate activities together and corporate to solve problems (Danese, 2013; Zhao et al., 2013).

Measurement model and construct unidimensionality, reliability and validity

To further validate the scales, I tested the measurement constructs and the complete measurement model in LISREL 8.80. This attempt requires the preliminary investigation of survey data in order to meet the normality assumptions of the covariance-based modelling software and to handle data limitations regarding non-responses to single questions (Hair, Black, Anderson & Tatham, 1998; Hair, Ringle & Sarstedt, 2011). In that sense, issues concerning incomplete data were mitigated first by replacing missing values with the expectation maximization (EM) method as suggested by Schumacker and Lomax (2012). As a second step, outliers were detected based on the work of Hoaglin and Iglewicz (1987) and Hoaglin et al. (1986). More precisely, values that were smaller or greater than 2.4 interquartile ranges from the upper and lower quantile respectively were la-beled as outliers and replaced by the mean of the variable (Tabachnick & Fidell, 2007). As a third step, kurtosis and skewness were investigated by inspecting Q-Q plots and homoscedasticity as well as linearity by reviewing scatterplots for each item (Hair et al., 1998). In both cases, univari-ate normality was ensured and no further concerns raised about the appropriunivari-ateness of the data.

Construct reliability. To further verify the appropriateness of each first-order construct, the reliability of each concept and its associated scales were reviewed in terms of stability, predicta-bility, accuracy, consistency and thus repeatability (Carmines & Zeller, 1979). In line with several studies in OM literature, an evidence of those aspects and, hence, of scale reliability is provided by presenting the composite reliability and Cronbach’s alpha of each factor (Blome, Schoenherr & Eckstein, 2014; D. X. Peng & Lai, 2012; Swafford, Ghosh & Murthy, 2008; Fornell & Larcker, 1981). As can be seen in Table 4, 5, 6 and 7, all constructs show values greater than 0.70 for both quality criteria and are therefore considered as internally consistent according to Bagozzi and Yi (1988) and Karlsson (2016).

Construct validity. Finally, to consider the measurement constructs further as valid, I verified that the intended and the actual measurement match; i.e. I prove that each item and upper-level construct is aligned with its associated theoretical concept and represents it accordingly (Karlsson, 2016). In agreement with Campbell and Fiske (1959), this proof of construct validity is given by the fulfillment of both convergent and discriminant validity.

Convergent validity refers to the extent to which the individual items of a measurement con-struct are aligned (Karlsson, 2016). A common practice to test this alignment is to perform a CFA (Bagozzi, Yi & Phillips, 1991) and evaluate whether factors satisfy two criteria. First, an item has to load significantly on its first-order construct and a first-order construct significantly on its sec-ond-order construct. Secondly, all factor loadings must be greater than 0.50 (Anderson & Bulletin, 1988). Table 4, 5, 6 and 7 shows that both criteria are met. All factor loadings are greater than the threshold value and all corresponding t-values are significant at the 0.01 level.

TABLE 4

Measurement scales assessment of Just-In-Time practice bundle

First-order construct Composite

reliability Cronbach’s alpha Indicator Factor loading t-value

Daily schedule adherence (DSA) 0.820 0.812 0.764* _13.018

J_DSA1 0.891 - J_DSA2 0.583 11.090 J_DSA3 0.856 18.466 J_DSA6 0.555 10.435 Flow-oriented layout (FL) 0.825 0.824 0.782* _11.049 J_FL1 0.724 - J_FL4 0.806 13.011 J_FL5 0.770 12.540 J_FL6 0.637 10.484

JIT link with suppliers (LWS) 0.744 0.748 0.878* _11.535

J_LWS1 0.699 - J_LWS2 0.564 8.815 J_LWS3 0.673 10.301 J_LWS4 0.574 8.947 J_LWS5 0.516 8.109 Kanban (KAN) 0.839 0.834 0.574* _8.145 J_KAN2 0.677 - J_KAN3 0.851 12.585 J_KAN4 0.855 12.600

Setup time reduction (SET) 0.736 0,738 0.918* _11.274

J_SET1 0.663 -

J_SET2 0.661 9.792

J_SET3 0.680 10.021

J_SET4 0.557 8.487

TABLE 5

Measurement scales assessment of Total Quality Management practice bundle

First-order construct Composite

reliability Cronbach’s alpha Indicator Factor loading t-value

Statistical process control (PRC) 0.897 0.893 0.709* _10.933

T_PRC2 0.843 - T_PRC3 0.883 20.085 T_PRC4 0.669 13.292 T_PRC5 0.903 20.724 Process feedback (PRF) 0.814 0.811 0.879* _12.259 T_PRF1 0.799 - T_PRF2 0.706 12.399 T_PRF3 0.631 10.967 T_PRF4 0.751 13.219

Customer involvement (CIN) 0.792 0.792 0.618* _8.209

T_CIN1 0.708 -

T_CIN3 0.699 10.450

T_CIN5 0.726 10.735

T_CIN6 0.658 9.965

Supplier quality involvement (SIN) 0.801 0.798 0.642* _8.346

T_SIN2 0.670 -

T_SIN4 0.592 9.033

T_SIN5 0.789 11.238

T_SIN6 0.773 11.124

TABLE 6

Measurement scales assessment of Infrastructure practice bundle

First-order construct Composite

reliability Cronbach’s alpha Indicator Factor loading t-value

Small group problem solving (SMG) 0.874 0.870 0.817* _9.834

I_SMG1 0.610 - I_SMG2 0.810 11.289 I_SMG3 0.819 11.368 I_SMG4 0.825 11.414 I_SMG5 0.605 9.127 I_SMG6 0.703 10.228

Multi-functional employees (MFE) 0.845 0.843 0.656* _9.700

I_MFE1 0.775 -

I_MFE2 0.846 14.953

I_MFE4 0.780 13.920

I_MFE5 0.628 10.983

Employee suggestion (EMS) 0.862 0.859 0.780* _11.638

I_EMS1 0.780 -

I_EMS2 0.800 14.880

I_EMS3 0.716 13.074

I_EMS4 0.843 15.776

I_EMS5 0.567 10.057

Management commitment (MCO) 0.856 0.853 0.665* _8.892

I_MCO1 0.653 -

I_MCO2 0.831 12.155

I_MCO4 0.640 9.899

I_MCO5 0.804 11.883

I_MCO6 0.746 11.233

TABLE 6 (CONTINUED)

First-order construct Composite

reliability Cronbach’s alpha Indicator Factor loading t-value Autonomous maintenance (TPM) 0.761 0.760 0.665* _7.748 I_TPM2 0.695 - I_TPM3 0.549 8.209 I_TPM5 0.717 10.053 I_TPM6 0.696 9.882

* Factor loading on the second-order construct

TABLE 7

Measurement scales assessment of moderators

First-order construct Composite

reliability Cronbach’s alpha Indicator Factor loading t-value Product modularity (PM) 0.749 0.735 MOD1 0.644 - MOD2 0.535 8.253 MOD3 0.904 9.458 MOD5 0.500 7.777

Internal integration (INT) 0.868 0.869

INT1 0.846 -

INT2 0.774 15.069

INT4 0.724 13.880

TABLE 8

Delta Chi-square test of first-order constructs

TABLE 9

Delta Chi-square test of first-order constructs and second-order constructs

JIT TQM INFRA PM INT INN

JIT - TQM 155.99 - INFRA 135.26 122.68 - PM 158.66 212.72 197.71 - INT 128.88 118.89 91.92 145.16 - INN 204.63 219.07 177.20 582.34 158,18 -

INFRA = Infrastructure: INN = Innovation performance

Model fit. Finally, the suitability of the whole measurement model, i.e. the general model fit, was assessed in two steps. Firstly, the second-order constructs were reviewed similar to Bortolotti, Danese et al. (2015). For each of the upper-level factors the underlying measurement model was elaborated in LISREL and evaluated accordingly. Table 10 shows that the fit statistics for all three practice bundles are acceptable. The threshold values of the absolute and incremental fit indices RMSEA and CFI are met and all normed chi-squares (χ2_{/df) are smaller than three (Hair, Black,}

Babin, Anderson & Tatham, 2006). In regard to the model fit of the second-order constructs, I simplified the models next in order to avoid an overparameterization and thus enhance the statis-tical power of the research (Gefen, Straub & Boudreau, 2000). I parceled the items of the first-order constructs by computing an average of these and, henceforth, measured the practices as an aggregate (McKone et al., 2001; Hall, Snell & Foust, 1999). Secondly, the whole measurement model was evaluated; including the second-order constructs along with the two first-order factors product modularity and internal integration. The assessment revealed acceptable fit indices:

χ2 _{= 467.08; d.f. = 238; χ}2 _{/d.f. = 1.96 < 3; RMSEA = 0.054 < 0.08; CFI = 0.970 > 0.90}

TABLE 10

Measurement models of second-order constructs

Second-order construct χ2 _{d.f. χ}2 _{/d.f. REMSEA CFI}

JIT 462.79 165 2.80 0.078 0.958

TQM 276.71 100 2.76 0.075 0.965

ANALYSIS AND RESEARCH RESULTS

Structural model

After above assessment of the measurement model, the theoretical framework presented in section 3 was tested by structural equation modelling (SEM) in LISREL 8.80. Taking up on the advantage of SEM, the technique allows to assess higher order models consisting of multiple exogenous and endogenous latent factors and to evaluate several relations simultaneously (Karlsson, 2016). Ac-cordingly, structural modelling enables the inclusion of multiple latent interaction variables, and besides, demonstrates a higher statistical power than moderated regression analysis. By applying the modelling method, researchers reduce the issue of biased estimates as SEM corrects for meas-urement errors while computing an approximation for the interaction effects (F. Li et al., 1998; Jaccard & Wan, 1995; Bagozzi & Yi, 1988). Given this characteristics, SEM is considered as most appropriate analysis method for the present research and also as accurate since it is applied as confirmatory analysis method as proposed by academia (Gefen et al., 2000).

With regard to the additional testing, academia reports a variety of different methods for the incorporation of interaction effects. Particularly multiple-group approaches are generally adapted in case of discrete moderators. Researchers assess interaction effects by evaluating several mod-els, each representing a possible value of the moderator, and interpret the results for each group (Cortina, Chen & Dunlap, 2001). Though these procedures are limited to categorial variables and the moderators observed in this research define the opposite. Both internal integration and product modularity are meant to classify a degree of adaption and achievement and not to categorize the plants of this research. Considering the opposite, Cortina et al. (2001) discusses six widely-used procedures for continuous variables and conclude that the various methods suggested in literature show similar results. However, the authors argue that the methods of Mathieu, Tannenbaum and Salas (1992) and Ping (1995) can be easier applied operationally and conceptually (Cortina et al., 2001). Hence, to take advantage of the ease and also to be consistent with other studies in man-agement literature (Bortolotti et al., 2013; Mesquita & Lazzarini, 2008; Devaraj, Krajewski & Wei, 2007), I applied the latter method and included both moderations guided by the procedure of Ping (1995).

factor loading:

𝜆IV_X_M= (𝜆IV1+ 𝜆IV2) ∗ (𝜆M1 + 𝜆M2)

measurement error (theta-delta):

𝜃IV_X_M

𝛿 _{= (𝜆}

IV1+ 𝜆IV2)2∗ 𝑉𝑎𝑟(𝐼𝑉) ∗ (𝜃IV1𝛿 + 𝜃IV2𝛿 ) + (𝜆M1 + 𝜆M2)2

∗ 𝑉𝑎𝑟(𝑀) ∗ (𝜃M1𝛿 + 𝜃M2𝛿) + (𝜃IV1𝛿 + 𝜃IV2𝛿 )

∗ (𝜃_M1𝛿 _{+ 𝜃}

M2𝛿)

By adapting the equations correspondingly, I calculated the loadings and measurement errors for each potential interaction effect of this research. In line with the second step of Ping (1995), the computation was followed by the elaboration of a structural model that incorporates the inter-action effects of interest. The effects were represented as a latent product, i.e. a single-item factor that is based on the underlying variables and calculated by

𝑥IV_X_M= (𝑥IV1+ 𝑥IV2) ∗ (𝑥M1 + 𝑥M2)

More precisely, each interaction effect is measured through a value that is computed by mul-tiplying the sum of the items associated with the independent variable with the sum of the items associated with the moderator. Additionally, its factor loading and error are fixed at the previously calculated parameter estimates. An aspect that shall not mitigate the statistical power of this study as all measurement constructs are unidimensional as demonstrated earlier (Cortina et al., 2001). Also, it is to be noted that the assumption of multicollinearity is not violated. All variance inflation indices (VFI) were computed and were below three (Hair et al., 1998). Figure 2 pictures the final structural model including the statistical significant results of the SEM analysis.

Hypothesized findings

The fit statistics of the final structural model demonstrate a good fit of the data:

χ2 _{= 677.08; d.f. = 376; χ}2 _{/d.f. = 1.80 < 3; RMSEA = 0.049 < 0.08; CFI = 0.965 > 0.90}

relationship was not introduced in the theoretical framework, but evaluated due to the application of Ping (1995). However, none of the other two practice bundles significantly affect innovation performance rejecting hypothesis 1 and 2.

With respect to the introduced interaction effects, the results are mixed. Whereas, product modularity does not moderate any of the lean-innovation relationships, internal integration deter-mines the impact of Total Quality Management and Infrastructure on innovation. Hence, statisti-cal evidence is given for hypothesis 7 (γ = 0.58; t-value = 2.67; p-value < 0.01) and hypothesis 8 (γ = -0.52; t-value = -2.26; p-value < 0.05), but it cannot be concluded that a reduction of product variety may enhance any relationship. Hence, hypothesis 4 to 6 are not hold. However, to provide a clear understanding of how internal integration impacts certain linkages, figure 3 and 4 visualize the interaction effects. The two plots picture the amount of innovation performance achieved by both lean bundles at a low and a high extend of internal integration, i.e. at one standard deviation below and one standard deviation above the mean as proposed in literature (Aiken, West & Reno, 1991). The underlying equations are given below the graphs respectively. Table 11 further sum-marizes the above-mentioned main and interaction effects to give an overall picture.

FIGURE 2

TABLE 11

Results of the structural model

Exogenous variable* Hypothesis Standardized coefficient t-value p-value

TQM H1 - 0.35 Not supported

INFRA H2 + 0.67 + 2.93 0.01

JIT H3 - 0.06 Not supported

INT - 0.03

PM + 0.19

TQM_X_PM H4 + 0.04 Not supported

INFRA_X_PM H5 + 0.06 Not supported

JIT_X_PM H6 - 0.06 Not supported

TQM_X_INT H7 + 0.58 + 2.67 0.01

INFRA_X_INT H8 - 0.52 - 2.26 0.05

JIT_X_INT H9 - 0.12 Not supported

* Endogenous variable equals innovation performance in all cases

FIGURE 3

Plot of interaction effect between Total Quality Management and Internal Integration

FIGURE 4

Plot of interaction effect between Infrastructure and Internal Integration

Equation given by INN = (0.65) * INFRA + ( – 0.03) * INT + ( – 0.52) * INFRA_X_INT

DISCUSSION

Theoretical implications

Main effects. The results provide several insights about the relationship between lean as a socio-technical production system and innovation as a capability to introduce innovative products in a timely manner. In general, previous research solely reports insights about diverse lean practices in relation to innovation within a single framework. This thesis provides such greater picture of this relationship by considering the four practices bundles introduced by Shah and Ward (2003): JIT and TQM as well as HRM and TPM. In particular, by viewing the two later practice bundles as a single construct, this work further expands the OM knowledge base about shared practices between JIT and TQM. Relating thereto, it is to be noted that each practice bundle is conceptual-ized as a second order construct. Thus the framework under study reflects not only multiple lean facets that are of interest in OM literature but also forms a comprehensive picture of lean manu-facturing similar to Bortolotti et al. (2015) since it captures all relevant, unobservable domains of each underlying practice and in turn of each practice bundle.

Concerning leans quality facets, the model further provides greater insights about the relation between quality management and innovation. The findings do not support the hypothesized pos-itive relationship between TQM and innovation performance; but instead suggest a negative as-sociation between those constructs. On the one hand, this is in contrasts with studies that empiri-cally determined a positive relationship (Singh & Smith, 2004; Prajogo & Sohal, 2003). On the other hand, however, it is consistent with the theoretical propositions of researchers that support the opposite field of research (Perdomo-Ortiz, González-Benito & Galende, 2009). The alignment with the later might be explained through above conceptualization of TQM. In contrast to other studies, this research considers soft quality management practices such as small group problem solving, management commitment and employee suggestions as part of the Infrastructure bundle. For instance, leadership expertise may encourage individuals to invent and improve current struc-tures, which in turn may lead to an advanced innovation performance. Also, this study considers partially technic-specific procedures as process feedback and control, that might slow down prod-uct innovation, as well as customer involvement which might result in solely taking an existing customer base into account and mitigating attempts of producing innovations (Prajogo & Sohal, 2004). Finally, it is important to note that by focusing on product innovation and speed of product introduction this research neglected particular facets of innovation which might be facilitated by TQM methods, e.g. incremental or process inventions.

such as multi-functional employees and small group problem solving encourage the personnel to question current processes and suggest improvements. It confirms that especially practices com-mon in both TQM and JIT management contexts relate to an enhanced innovation performance. Further, in contrast to earlier publications in OM literature, this study specifically examines JIT practices in relation to multiple facets of a firm’s innovation expertise. In terms of how quick products are introduced into the market, prior JIT-related research is limited to the investigation of one inventive capability, i.e. fast product innovation (cf. Matsui, 2007; Flynn & Flynn, 2004). Instead this research considers product innovativeness as second facet and examined the relation-ship between JIT and innovation as multi-item scale performance dimension. As shown in table 11, the analysis revealed that the practice bundle has a non-significant, and almost non-directional effect on such an innovation construct. This implies that JIT does not hinder innovation. Alt-hough, JIT may not facilitate innovativeness itself due to its characteristic of working best in repetitive and standardized production systems (Hopp & Spearman, 2008), it may indirectly sup-port the speed of new product introductions. Based on its indirect relation to flexibility perfor-mance (Bortolotti, Danese, et al., 2015), JIT practices may improve the ability to adapt a produc-tion system to new products and thus support fast product innovaproduc-tion. In fact, with the purpose of a lot size of one through the continuous reduction of setup times, JIT aims for the maximization of responsiveness and thereby allows for a quick inclusion of products into the master production schedule (Hopp & Spearman, 2008). Also, the implementation of a JIT link with suppliers might enhance a firm’s capability to introduce products more quickly as enhanced relationships between buyer and supplier might simplify the procurement of components that are required.

Interaction effects. Furthermore, this work gives an essential understanding about the con-textual environment in which lean may enhance a firm’s innovation capabilities best. Given the results, the first moderator in terms of the conceptual decomposition and standardization of prod-ucts does not enhance any observed lean-innovation relationship. However, the second moderator under study strengthens and weakens two relations. Firstly, internal integration shows to have a positive impact on the TQM-innovation relationship. Quality management attempts become more successful when a firm’s various functions corporate and share information. Cross-functional in-tegration aid in the improved inin-tegration of customer and supplier (Alfalla-Luque, Marin-Garcia & Medina-Lopez, 2015) and thus significantly weakens the negative impact of those quality prac-tices considered in this study.

groups. Individual, smaller groups may promote innovative ideas since groups are enabled to take ownership and feel free to experiment and create new solutions (Kim & Mauborgne, 1999).

Managerial implications

Finally, it is to be noted that the findings have essential propositions that should be of concern for executives hired not only for shop floor positions but also for administrative functions. Bearing the specific innovation facets under study in mind, such proposals are particularly of interest by means of how to enhance a company’s product innovativeness as well as its speed of introducing new goods to the market. Thus, this work is to be interpreted in the light of such two aspects and be especially of interest for optimizing product innovation.

To begin with the direct effects of lean manufacturing on performance, executives operating at either of the two hierarchical levels are advised to facilitate the implementation of HRM prac-tices. Indeed, the significant impact of the infrastructure bundle suggests managers to support the individual. Through attempts of cross-training and encouraging employees to solve problems in-dividually and in teams, an organization enhances entrepreneurial creativity and in turn promotes in-house innovation. As a result, the analysis further indicates it is essential that managers provide leadership and pay attention to ideas of the individual and acknowledge suggestions as those help to improve innovation performance significantly.

Despite, this thesis developed an understanding of how TQM and JIT affect innovation per-formance as set forth in the previous section. The results provide an argumentation against a non- beneficial role of lean in relation to innovation. However, the results indicate a negative impact of TQM, the effect of TQM is not significant and thus not to be seen as considerable. More im-portantly, this work provides empirical evidence that JIT does not negatively impacts a company’s innovation capability. Hence, with the positive impact of each infrastructure practice in mind, it is recommended to adapt lean as a set of management practices as it influences innovation as a general.

CONCLUSIONS

The purpose of this research was to contribute to the mature knowledge base of lean manufactur-ing by simultaneously evaluatmanufactur-ing relationships between multiple practice bundles and innovation performance as well as examining the impact of two contextual factors on such linkages. Previous discussion of those attempts validates that lean as a whole does not hinder innovation but instead encourages it. Especially, by means of the observed infrastructure practices, the ‘soft’ part of lean positively affects innovativeness. Moreover, the findings reveal that internal integration defines a twofold moderator and thus a key trigger. Whereas it positively moderates the link of TQM and innovation, it weakens the impact of infrastructure practices on innovation. It emphasizes that managers need to acknowledge the role of entrepreneurial individualism in the process of shaping innovation and need to balance the degree of integration for each bundle separately.

As a final note, it has to be recalled that every research is not without limitations. First, I used perceptual data only in this thesis. However, this procedure is commonly used within OM litera-ture, it may be argued that a combination of objective and subjective measures, especially in case of the endogenous variable, is more appropriate. Future studies may complement perceptual data and overcome the disadvantages linked to it.

Second, this study was conducted at the plant level of manufacturers competing with multina-tional and world-class companies and also operating with more than 100 employees. This implies pre-defined limitations of generalizability but also opportunities for future research. On the one hand, the study should be reproduced for non-manufacturers and so in a different industry setting. A study conducted at an organizational-level in the service may reveal different results. Since the differentiation between technical and socio may diminish and the theoretical association of JIT practices with standardization at the shop floor level may change, observations will provide new insights. On the other hand, the repetition of the study for micro and small companies defines a fruitful research field since a successful implementation of lean is often associated with a certain degree of resource power and company size.

Third, future research should be carried out by using different techniques for evaluating mod-eration effects. The method used in this study does not allow for the concurrent evaluation of mediation and moderation. However, given that infrastructure practices act as antecedents for JIT and TQM, it is worthy to understand how this linkage affect the observations of this study.

R E F E R E N C E S

Ahmad, S., & Schroeder, R. G. 2003. The impact of human resource management practices on operational performance: recognizing country and industry differences. Journal of Operations Management, 21(1): 19–43.

Ahmad, S., Schroeder, R. G., & Sinha, K. K. 2003. The role of infrastructure practices in the effectiveness of JIT practices: implications for plant competitiveness. Journal of Engineering and Technology Management, 20(3): 161–191.

Aiken, L. S., West, S. G., & Reno, R. R. 1991. Multiple regression: Testing and interpreting interactions. Sage Publications.

Alfalla-Luque, R., Marin-Garcia, J. A., & Medina-Lopez, C. 2015. An analysis of the direct and mediated effects of employee commitment and supply chain integration on organisational performance. International Journal of Production Economics, 162: 242–257.

Anderson, J. C., & Bulletin, D. W. 1988. Structural Equation Modeling in Practice: A Review and Recommended Two-Step Approach. Psychological Bulletin, 103(3): 411–423. Araujo, L., Dubois, A., & Gadde, L.-E. 1999. Managing Interfaces with Suppliers. Industrial

Marketing Management, 28(5): 497–506.

Bagozzi, R. P., & Yi, Y. 1988. On the evaluation of structural equation models. Journal of the Academy of Marketing Science, 16(1): 74–94.

Bagozzi, R. P., Yi, Y., & Phillips, L. W. 1991. Assessing Construct Validity in Organizational Research. Administrative Science Quarterly, 36(3): 421.

Bevilacqua, M., Ciarapica, F. E., & De Sanctis, I. 2017. Lean practices implementation and their relationships with operational responsiveness and company performance: an Italian study. International Journal of Production Research, 55(3): 769–794.

Blome, C., Schoenherr, T., & Eckstein, D. 2014. The impact of knowledge transfer and complexity on supply chain flexibility: A knowledge-based view. International Journal of Production Economics, 147: 307–316.

Bortolotti, T., Boscari, S., & Danese, P. 2015. Successful lean implementation: Organizational culture and soft lean practices. International Journal of Production Economics,

Bortolotti, T., Danese, P., Flynn, B. B., & Romano, P. 2015. Leveraging fitness and lean bundles to build the cumulative performance sand cone model. International Journal of Production Economics, 162: 227–241.

Bortolotti, T., Danese, P., & Romano, P. 2013. Assessing the impact of just-in-time on operational performance at varying degrees of repetitiveness. International Journal of Production Research, 51(4): 1117–1130.

Boyer, K. K., & Lewis, W. 2002. Competitive priorities: Investigating the need for trade-offs in operations strategy. Productions and operations management, 11(1): 9–20.

Bozarth, C. C., Warsing, D. P., Flynn, B. B., & Flynn, E. J. 2009. The impact of supply chain complexity on manufacturing plant performance. Journal of Operations Management, 27(1): 78–93.

Brown, S., Squire, B., & Blackmon, K. 2007. The contribution of manufacturing strategy involvement and alignment to world‐class manufacturing performance. International Journal of Operations & Production Management, 27(3): 282–302.

Campbell, D. T., & Fiske, D. W. 1959. Convergent and discriminant validation by the multitrait-multimethod matrix. Psychological Bulletin, 56(2): 81–105.

Carmines, E., & Zeller, R. 1979. Reliability and validity assessment, Vol. 17. Sage publications.

Chowdhury, M., Jayaram, J., & Prajogo, D. 2017. The influence of socialisation and absorptive capacity on buyer’s innovation performance. International Journal of Production Research, 55(23): 7022–7039.

Cortina, J. M., Chen, G., & Dunlap, W. P. 2001. Testing Interaction Effects in LISREL:

Examination and Illustration of Available Procedures. Organizational Research Methods, 4(4): 324–360.

Cua, K. O., McKone, K. E., & Schroeder, R. G. 2001. Relationships between implementation of TQM, JIT, and TPM and manufacturing performance. Journal of Operations

Management, 19(6): 675–694.

Danese, P. 2013. Supplier integration and company performance: A configurational view. Omega, 41(6): 1029–1041.

Danese, P., & Filippini, R. 2010. Modularity and the impact on new product development time performance. International Journal of Operations & Production Management,

30(11): 1191–1209.

Danese, P., & Filippini, R. 2013. Direct and Mediated Effects of Product Modularity on Development Time and Product Performance. IEEE Transactions on Engineering Management, 60(2): 260–271.

Danese, P., & Romano, P. 2004. Improving inter‐functional coordination to face high product variety and frequent modifications. International Journal of Operations & Production Management, 24(9): 863–885.

Danese, P., Romano, P., & Bortolotti, T. 2012. JIT production, JIT supply and performance: investigating the moderating effects. Industrial Management & Data Systems, 112(3): 441–465.

Devaraj, S., Hollingworth, D. G., & Schroeder, R. G. 2001. Generic manufacturing strategies: an empirical test of two configurational typologies. Journal of Operations Management, 19(4): 427–452.

Devaraj, S., Krajewski, L., & Wei, J. C. 2007. Impact of eBusiness technologies on operational performance: The role of production information integration in the supply chain. Journal of Operations Management, 25(6): 1199–1216.

Ettlie, J. E. 1995. Product-Process Development Integration in Manufacturing. Management Science, 41(7): 1224–1237.

Farooq, M. A., Kirchain, R., Novoa, H., & Araujo, A. 2017. Cost of quality: Evaluating cost-quality trade-offs for inspection strategies of manufacturing processes. International Journal of Production Economics, 188: 156–166.

Feitzinger, E., & Lee, H. L. 1997. Mass Customization at Hewlett-Packard: The Power of Postponement. Harvard Business Review, 75: 116–123.

Fisher, M., Ramdas, K., & Ulrich, K. 1999. Component Sharing in the Management of Product Variety: A Study of Automotive Braking Systems. Management Science.

Flynn *, B. B., & Flynn, E. J. 2005. Synergies between supply chain management and quality management: emerging implications. International Journal of Production Research, 43(16): 3421–3436.

Flynn, B. B., & Flynn, E. J. 1999. Information-Processing Alternatives for Coping with Manufacturing Environment Complexity. Decision Sciences, 30(4): 1021–1052.

Flynn, B. B., & Flynn, E. J. 2004. An exploratory study of the nature of cumulative capabilities. Journal of Operations Management, 22(5): 439–457.

Flynn, B. B., Sakakibara, S., & Schroeder, R. G. 1995. Relationship Between JIT and TQM: Practices and Performance. Academy of Management Journal, 38(5): 1325–1360. Flynn, B. B., Schroeder, R. G., & Sakakibara, S. 1995. The Impact of Quality Management

Practices on Performance and Competitive Advantage. Decision Sciences, 26(5): 659–691.

Fornell, C., & Larcker, D. F. 1981. Evaluating Structural Equation Models with Unobservable Variables and Measurement Error. Journal of Marketing Research, 18(1): 39–50. Forza, C., Salvador, F., & Simionato, T. 2000. An empirical study on the efficiency and

effectiveness of new product development in high product variety environments. Working paper no. 009-2000. Padua, IT.

Furlan, A., Vinelli, A., & Dal Pont, G. 2011. Complementarity and lean manufacturing bundles: an empirical analysis. International Journal of Operations & Production Management, 31(8): 835–850.

Gefen, D., Straub, D., & Boudreau, M.-C. 2000. Communications of the Association for Information Systems Structural Equation Modeling and Regression: Guidelines for Research Practice. Communications of the Association for Information Systems, 4(1): 7. Hair, J. F., Black, W. C., Anderson, R. E., & Tatham, R. L. 1998. Multivariate data analysis.

Prentice-Hall.

Hair, J. F., Ringle, C. M., & Sarstedt, M. 2011. PLS-SEM: Indeed a Silver Bullet. The Journal of Marketing Theory and Practice, 19(2): 139–152.

Hall, R. J., Snell, A. F., & Foust, M. S. 1999. Item Parceling Strategies in SEM: Investigating the Subtle Effects of Unmodeled Secondary Constructs. Organizational Research Methods, 2(3): 233–256.

Hallgren, M., & Olhager, J. 2009. Flexibility configurations: Empirical analysis of volume and product mix flexibility. Omega, 37(4): 746–756.

Heim, G. R., & Peng, D. X. 2010. The impact of information technology use on plant structure, practices, and performance: An exploratory study. Journal of Operations Management, 28(2): 144–162.

Hoaglin, D. C., & Iglewicz, B. 1987. Fine-Tuning Some Resistant Rules for Outlier Labeling. Journal of the American Statistical Association, 82(400): 1147–1149.

Hoaglin, D. C., Iglewicz, B., & Tukey, J. W. 1986. Performance of Some Resistant Rules for Outlier Labeling. Journal of the American Statistical Association, 81(396): 991–999. Hopp, W. J. 2011. Supply chain science. Waveland Press.

Hopp, W. J., & Spearman, M. L. 2008. Factory Physics: foundation of manufacturing management. Waveland Press.

Hu, L., & Bentler, P. M. 1999. Cutoff criteria for fit indexes in covariance structure analysis: Conventional criteria versus new alternatives. Structural Equation Modeling:

A Multidisciplinary Journal, 6(1): 1–55.

Huang, X., Kristal, M. M., & Schroeder, R. G. 2008. Linking learning and effective process implementation to mass customization capability. Journal of Operations Management, 26(6): 714–729.

Hung, R. Y. Y., Lien, B. Y.-H., Yang, B., Wu, C.-M., & Kuo, Y.-M. 2011. Impact of TQM and organizational learning on innovation performance in the high-tech industry. International Business Review, 20(2): 213–225.

Jaccard, J., & Wan, C. K. 1995. Measurement Error in the Analysis of Interaction Effects Between Continuous Predictors Using Multiple Regression: Multiple Indicator and Structural Equation Approaches. Psychological Bulletin, 117(2): 348–357.

Jacobs, M., Droge, C., Vickery, S. K., & Calantone, R. 2011. Product and Process Modularity’s Effects on Manufacturing Agility and Firm Growth Performance. Journal of Product Innovation Management, 28(1): 123–137.

Jacobs, M., Vickery, S. K., & Droge, C. 2007. The effects of product modularity on competitive performance. International Journal of Operations & Production Management,

27(10): 1046–1068.

Jayaram, J., Ahire, S. L., & Dreyfus, P. 2010. Contingency relationships of firm size, TQM duration, unionization, and industry context on TQM implementation—A focus on total effects. Journal of Operations Management, 28(4): 345–356.

Kanji, G., & Asher, M. 1996. 100 methods for total quality management. Sage.

Karlsson, C. 2016. Research methods for operations management. New York: Routledge. Kaynak, H. 2003. The relationship between total quality management practices and their effects

on firm performance. Journal of Operations Management, 21(4): 405–435.

Ketokivi, M. A., & Schroeder, R. G. 2004. Strategic, structural contingency and institutional explanations in the adoption of innovative manufacturing practices. Journal of Operations Management, 22(1): 63–89.

Kim, W. C., & Mauborgne, R. 1999. Strategy, value innovation, and the knowledge economy. Sloan Management Review, 40(3): 41–42.

Lander, E., & Liker, J. K. 2007. The Toyota Production System and art: making highly customized and creative products the Toyota way. International Journal of Production Research, 45(16): 3681–3698.

Lau, A. K. W., Yam, R. C. M., & Tang, E. 2011. The Impact of Product Modularity on New Product Performance: Mediation by Product Innovativeness. Journal of Product Innovation Management, 28(2): 270–284.

Li, F., Harmer, P., Duncan, T. E., Duncan, S. C., Acock, A., & Boles, S. 1998. Approaches to Testing Interaction Effects Using Structural Equation Modeling Methodology.

Multivariate Behavioral Research, 33(1): 1–39.

Li, S., Rao, S. S., Ragu-Nathan, T. S., & Ragu-Nathan, B. 2005. Development and validation of a measurement instrument for studying supply chain management practices. Journal of Operations Management, 23(6): 618–641.

Liu, G. J., Shah, R., & Schroeder, R. G. 2006. Linking Work Design to Mass Customization: A Sociotechnical Systems Perspective. Decision Sciences, 37(4): 519–545.

Maccallum, R. C., Browne, M. W., & Sugawara, H. M. 1996. Power Analysis and Determination of Sample Size for Covariance Structure Modeling. Psychological Methods, (2): 130–149.

Mackelprang, A. W., & Nair, A. 2010. Relationship between just-in-time manufacturing practices and performance: A meta-analytic investigation. Journal of Operations Management, 28(4): 283–302.

Mathieu, J. E., Tannenbaum, S. I., & Salas, E. 1992. Influences of Individual and Situational Characteristics on Measures of Training Effectiveness. Academy of Management Journal, 35(4): 828–847.

Matsui, Y. 2007. An empirical analysis of just-in-time production in Japanese manufacturing companies. International Journal of Production Economics, 108(1–2): 153–164. McKone, K. E., Schroeder, R. G., & Cua, K. O. 2001. The impact of total productive

maintenance practices on manufacturing performance. Journal of Operations Management, 19(1): 39–58.

Mesquita, L. F., & Lazzarini, S. G. 2008. Horizontal and Vertical Relationships in Developing Economies: Implications for SMEs’ Access to Global Markets. Academy of Management Journal, 51(2): 359–380.

Mikkola, J. H., & Gassmann, O. 2003. Managing modularity of product architectures: toward an integrated theory. IEEE Transactions on Engineering Management, 50(2): 204–218. Nair, A. 2006. Meta-analysis of the relationship between quality management practices and firm

Narasimhan, R., & Das, A. 2001. The impact of purchasing integration and practices on manufacturing performance. Journal of Operations Management, 19(5): 593–609. Narasimhan, R., & Schoenherr, T. 2013. Revisiting the progression of competitive capabilities:

results from a repeated cross-sectional investigation. International Journal of Production Research, 51(22): 6631–6650.

Narasimhan, R., Swink, M., & Kim, S. W. 2006. Disentangling leanness and agility: An empirical investigation. Journal of Operations Management, 24(5): 440–457.

Netland, T. H., Schloetzer, J. D., & Ferdows, K. 2015. Implementing corporate lean programs: The effect of management control practices. Journal of Operations Management, 36: 90–102.

Noble, M. A. 1995. Manufacturing Strategy: Testing the Cumulative Model in a Multiple Country Context. Decision Sciences, 26(5): 693–721.

O’Leary-Kelly, S. W., & J. Vokurka, R. 1998. The empirical assessment of construct validity. Journal of Operations Management, 16(4): 387–405.

Peng, D. X., & Lai, F. 2012. Using partial least squares in operations management research: A practical guideline and summary of past research. Journal of Operations Management, 30(6): 467–480.

Peng, D. X., Schroeder, R. G., & Shah, R. 2008. Linking routines to operations capabilities: A new perspective. Journal of Operations Management, 26(6): 730–748.

Peng, X. D., Liu, G. J., & Heim, G. R. 2011. Impacts of information technology on mass customization capability of manufacturing plants. International Journal of Operations & Production Management, 31(10): 1022–1047.

Perdomo-Ortiz, J., González-Benito, J., & Galende, J. 2009. The intervening effect of business innovation capability on the relationship between Total Quality Management and

technological innovation. International Journal of Production Research, 47(18): 5087–5107.

Ping, R. A. 1995. A Parsimonious Estimating Technique for Interaction and Quadratic Latent Variables. Journal of Marketing Research, 32(3): 336.

Powell, T. C. 1995. Total quality management as competitive advantage: A review and empirical study. Strategic Management Journal, 16(1): 15–37.

Prajogo, D. I., & Hong, S. W. 2008. The effect of TQM on performance in R&amp;D

environments: A perspective from South Korean firms. Technovation, 28(12): 855–863. Prajogo, D. I., & Sohal, A. S. 2003. The relationship between TQM practices, quality

performance, and innovation performance. International Journal of Quality & Reliability Management, 20(8): 901–918.

Prajogo, D. I., & Sohal, A. S. 2004. The multidimensionality of TQM practices in determining quality and innovation performance — an empirical examination. Technovation, 24(6): 443–453.

Roemer, T. A., Ahmadi, R., & Wang, R. H. 2000. Time-Cost Trade-Offs in Overlapped Product Development. Operations Research, 48(6): 858–865.

Sadikoglu, E., & Zehir, C. 2010. Investigating the effects of innovation and employee performance on the relationship between total quality management practices and firm performance: An empirical study of Turkish firms. International Journal of Production Economics, 127(1): 13–26.

Safizadeh, M. H., Ritzman, L. P., & Mallick, D. 2000. Revisiting alternative theoretical paradigms in manufacturing strategy. Production and Operations Management, 9(2): 111–126.

Safizadeh, M. H., Ritzman, L. P., Sharma, D., & Wood, C. 1996. An Empirical Analysis of the Product-Process Matrix. Management Science, 42(11): 1576–1591.

Sakakibara, S., Flynn, B. B., & Schroeder, R. G. 1993. A framework and measurement instrument for just‐in‐time manufacturing. Production and Operations Management, 2(3): 177–194.

Sakakibara, S., Flynn, B. B., Schroeder, R. G., & Morris, W. T. 1997. The Impact of Just-in-Time Manufacturing and Its Infrastructure on Manufacturing Performance. Management Science, 43(9): 1246–1257.