Performance effects of using an ERP system for ...

Journal of Operations Management 36 (2015) 147–164

Contents lists available at ScienceDirect

Journal of Operations Management journal homepage: www.elsevier.com/locate/jom

Performance effects of using an ERP system for manufacturing planning and control under dynamic market requirements Antti Tenhiälä a,∗ , Pekka Helkiö b,1 a b

IE Business School, Calle de María de Molina 12-5, 28006 Madrid, Spain Aalto University, Department of Industrial Engineering and Management, Otaniementie 17, 02150 Espoo, Finland

a r t i c l e

i n f o

Article history: Received 19 October 2012 Received in revised form 8 May 2014 Accepted 13 May 2014 Available online 21 May 2014 Keywords: Decision support systems Production scheduling Shop-floor control Materials management Strong inference Dynamic capabilities

a b s t r a c t Enterprise resource planning (ERP) systems have a controversial reputation. Critics say that even if ERP systems may be beneficial for organizations operating in stable conditions, they are surely detrimental to organizations that face dynamic market requirements. This is because ERP systems are said to impose such procedures and constraints on organizations that make business processes inflexible to change. In contrast, proponents argue that the information-processing capabilities of ERP systems are crucial for organizations that face dynamic market requirements and also that the criticized procedures and constraints actually support process reengineering. These two contradictory arguments are often found in practitioner literature, but both of them can also be supported by management theory. The central tenets of the Organic Theory of organization design imply that ERP systems should be detrimental when market requirements change frequently, whereas the principles of Rigid Flexibility Theory suggest that they should be advantageous. In this study, we use cross-sectional data from 151 manufacturing plants to determine which argument is more applicable in the context of manufacturing planning and control. The results strongly favor the use of ERP systems under dynamic market requirements. To facilitate the reconciliation of the two contradictory arguments, we discuss how the results may have been influenced by two contextual factors: the predominantly technical nature of the studied organizational system and the tight interdependence of the studied activities. © 2014 Elsevier B.V. All rights reserved.

1. Introduction Enterprise resource planning (ERP) systems attracted considerable attention in the business world at the turn of the millennium. Despite the sizable investments involved, these software packages were a breakthrough, especially in developed countries where they were installed by a vast majority of manufacturing firms (Olhager and Selldin, 2003; Jutras, 2010). After the initial excitement, however, an increasing number of managers have started to complain about the shortcomings of these systems. The main critique is that ERP systems impede making changes to business processes, which is a major problem in dynamic business environments where market requirements change rapidly (Rettig, 2007; Lindley et al., 2008; Goodhue et al., 2009; Ganly and Montgomery, 2012; Fauscette,

∗ Corresponding author. Tel.: +34 91 568 9600. E-mail addresses: [email protected] (A. Tenhiälä), pekka.helkio@aalto.fi (P. Helkiö). 1 Tel.: +358 9 47001. http://dx.doi.org/10.1016/j.jom.2014.05.001 0272-6963/© 2014 Elsevier B.V. All rights reserved.

2013; IDG Market Pulse, 2013). Frustrated with the inflexibility of ERP systems, many managers have sought help from in-house developed software or traditional functionally specialized business applications (Upton and Staats, 2008; Deloitte, 2010; Prouty and Castellina, 2011; Ganly and Montgomery, 2012). Often, if the managers have not taken the initiative to replace the ERP system, their subordinates have begun to circumvent its use (Bendoly and Cotteleer, 2008; Xue et al., 2011; Christiansen et al., 2012). Advocates of ERP systems abhor the implementation of standalone software as much as the circumvention tactics and claim that ERP systems can and should be always reconfigured when organizations change their business processes to serve new market requirements (Gattiker et al., 2005; Goodhue et al., 2009; Drobik and Rayner, 2011). They argue that the use of standalone tools or handcrafted spreadsheets compromises the main advantage of ERP systems, namely, the cross-functional integration that enables swift and reliable information flows across the organization (Berente and Yoo, 2012; Michael et al., 2012). The proponents of ERP systems do not perceive the alleged inflexibility as an obstacle but instead argue that the kind of rigidity that is inherent to ERP

148

A. Tenhiälä, P. Helkiö / Journal of Operations Management 36 (2015) 147–164

systems actually supports process reengineering and is therefore beneficial for organizations facing dynamic market requirements (Gattiker et al., 2005; Scott, 2011). Management theory helps understand the controversy among practitioners over whether ERP systems facilitate or impede adaptation to changing market requirements. We use Teece’s (2007) framework of dynamic capabilities to analyze the existing research and find out that ERP systems indeed have characteristics that both help and hinder organizational responses to dynamic market requirements. We then show that the conflicting arguments about the effects of ERP systems represent two different theoretical views on organizational adaptability. The argument that organizations facing dynamic market requirements must decentralize their management structures and rid themselves of the rules and standardized processes of ERP systems can be based on the Organic Theory of organization design (Burns and Stalker, 1961). The argument that these organizations benefit from the simplicity and discipline enforced by ERP systems can be based on Rigid Flexibility Theory (Collins and Schmenner, 1993). Although Organic Theory was first proposed over half a century ago, it has been widely used in the contemporary operations management research (e.g., Huang et al., 2010; Goodale et al., 2011; Zhang et al., 2012). Rigid Flexibility Theory is newer and less cited, but it too has received support in operations management research (Collins et al., 1998; da Silveira, 2006). Due to the theoretical backing for both perspectives, the dispute on whether ERP systems are beneficial or detrimental under dynamic market requirements cannot be settled via theoretical reasoning alone. To address the question empirically, we analyze data from 151 manufacturing plants in 12 industry sectors, focusing specifically on the performance of manufacturing planning and control activities under varying levels of dynamism in market requirements. We use Platt’s (1964) method of strong inference to test which of the two contradictory views is better supported in this empirical context. Finally, we discuss the theoretical generalizability of our findings by exploring the boundary conditions that may have influenced the results (Dubin, 1978). 2. Literature review To position the present study within the existing literature, we next discuss ERP systems’ status in today’s enterprise software landscape and then explore the question of how ERP systems may help or hinder operational performance when organizations face dynamic market requirements. 2.1. ERP systems in the enterprise software landscape of the 2010s ERP systems are modular software packages that integrate a firm’s business functions around a common database and standardized processes that are configured to fit the needs of the user organizations (e.g., Boudreau and Robey, 2005; Ranganathan and Brown, 2006; Sasidharan et al., 2012). Substantial research efforts have been directed to this special category of enterprise software. The notorious failures of some of the early implementations (see examples in, e.g., Robey et al., 2002) gave rise to much research on the typical pitfalls and success factors of implementing ERP systems (for reviews and recent examples, see, e.g., Karimi et al., 2007; Seddon et al., 2010; Berente and Yoo, 2012; Sasidharan et al., 2012; Yeh and Xu, 2013). The variability in the outcomes of the implementations also motivated broad research on ERP systems’ overall performance effects (e.g., Gattiker and Goodhue, 2005; Harris and Davenport, 2006; Ranganathan and Brown, 2006; Hendricks et al., 2007) and end users’ assimilation of the implemented systems (e.g., Boudreau and Robey, 2005; Bendoly and Cotteleer, 2008; Saeed

et al., 2010; Xue et al., 2011). A recurring finding in these studies has been that despite many successful examples (e.g., Cotteleer and Bendoly, 2006), the average overall performance effect of ERP systems has been fairly neutral (Bendoly et al., 2009), and that the variability in the effects cannot be fully explained with the implementation characteristics or the quality of use (Seddon et al., 2010). In response, the broad main streams of ERP system research have recently been complemented by studies that focus on specific contexts (e.g., Sarker et al., 2012; Lai et al., 2013) or pay special attention to ERP systems’ fitness to external contingencies (e.g., Berente and Yoo, 2012; Sasidharan et al., 2012). This study extends both of these emerging trends. Over time, the studies on ERP systems have also started to cover more specialized tools, such as customer relationship management, manufacturing planning, advanced production scheduling, supply chain management, and sourcing software (Hendricks et al., 2007; Sia and Soh, 2007; Stratman, 2007; Bendoly et al., 2008; Rai and Hornyak, 2013). This trend has been motivated by practical considerations, as firms have increasingly implemented standalone software (Deloitte, 2010; Fauscette, 2013). Although most firms do not have plans to completely abandon their ERP systems, many have implemented standalone tools to replace some functionality of their ERP systems (Ganly and Montgomery, 2012). Market research corroborates this trend by showing that the adoption rates of ERP system modules other than for financials and customer order handling have dropped considerably (Wailgum, 2008; Panorama Consulting, 2011) from the heyday of ERP systems in the early 2000s (Olhager and Selldin, 2003). Standalone tools are implemented especially in business functions where fitness to a firm’s processes is critical, a good example being manufacturing planning and control (Brandl, 2011). The trend in the increased use of standalone tools is interesting because it can be argued to compromise ERP systems’ key value offering: swift and reliable intra-organizational information flows (e.g., Michael et al., 2012). Whenever some information is managed outside the database of the ERP system, the integrity and currency of the information are put in jeopardy, regardless of whether rudimentary spreadsheets or sophisticated business applications are used (Berente and Yoo, 2012). The trend is also interesting because research has shown that wider functional scope would be associated with greater benefits from ERP systems (Ranganathan and Brown, 2006; Karimi et al., 2007). These contradictions motivate studying the question of whether replacing ERP system functionality with other tools is good or bad for operational performance. 2.2. Dynamic capabilities perspective on ERP systems Because ERP systems’ inflexibility is the most often cited reason why managers consider replacing them with other solutions (Upton and Staats, 2008; Goodhue et al., 2009; Ganly and Montgomery, 2012), this study explores whether the dynamism of the operating environment influences the performance effects of ERP systems. In the management literature, dynamism traditionally refers to the rate of change in those aspects of an organization’s environment that are not directly under its own control (Miller and Friesen, 1983). We focus on the rate of change in market requirements because it constitutes a great organizational challenge regardless of whether an ERP system or other software is used (Goodhue et al., 2009), as well as because research indicates that appropriate software is crucial to ensure “that firms can rapidly redesign existing processes and create new processes for exploiting dynamic marketplace conditions” (Sambamurthy et al., 2003, p. 245). This ability has been discussed as “market capitalizing agility” in the information systems literature (Lu and Ramamurthy, 2011) and under the rubric of “dynamic capabilities” in the management literature (Teece et al., 1997), yet it is not clear what kind of


software is appropriate for organizations that face dynamic market requirements. The concept of dynamic capabilities is fitting for the assessment of how ERP systems may facilitate or impede the adaptation of business processes to changing market requirements. That is because the dynamic capabilities refer to the activities “through which the organization systematically generates and modifies its operating routines” (Zollo and Winter, 2002, p. 340, where “operating routines” equate to “processes” in the operations management jargon), thus enabling organizations “to address rapidly changing environments” (Teece et al., 1997, p. 516). To analyze the potential advantages and disadvantages of ERP systems, we use Teece’s (2007) framework that divides dynamic capabilities into three capacities: (i) sensing new opportunities in the changing environment, (ii) seizing the identified opportunities, and (iii) transforming to better exploit the new opportunities. The literature suggests that ERP systems can affect all of them and that the effects can be both positive and negative. Most discussion on ERP systems’ advantages relates to the sensing capacity. The integrated processes, centralized databases, and automatic messaging features of ERP systems ensure that information generated in one part of an organization is available to the other parts without delays, loss of content, or distortions caused by interfaces between separate software tools (e.g., Bendoly and Schoenherr, 2005; Gattiker, 2007; Michael et al., 2012). In an integrated system, information about the changes in market conditions are quickly distributed across the organization (Gupta and Kohli, 2006). The automatic messaging features convey the implications of the changes as “exception messages” to all relevant decision makers based on the rules configured to the system (Jacobs et al., 2011; Tenhiälä and Salvador, 2014). The central database provides ample information for analyses on how to best respond to the observed changes (Setia and Patel, 2013). While necessary, the sensing capacity is not enough for organizations to prevail in dynamic market conditions. Organizations must also be able to adapt to the observed changes—which is the point of the seizing capacity of Teece’s (2007) framework. This is where the literature on ERP systems becomes mixed. In principle, all processes managed in ERP systems can be changed by reconfiguring the rules that control them, which is an activity supported by the systems’ built-in procedures (Gattiker et al., 2005; Goodhue et al., 2009; Drobik and Rayner, 2011; Scott, 2011). Yet, in practice, the reconfiguration work is generally considered onerous, as one must pay attention to the numerous interactions between the integrated features that are used across the organization, a fact that has led many firms to avoid changes as much as possible (Dechow and Mouritsen, 2005; Nash, 2010). Moreover, all changes must fit the data structures and the logic of the software, limiting the options for how processes can be redesigned and meaning that a perfect fit to the changed conditions is often unachievable (Sia and Soh, 2007; Strong and Volkoff, 2010). To free themselves from these constraints, some organizations have chosen to manage some or all of their processes without an ERP system (Upton and Staats, 2008; Deloitte, 2010; Prouty and Castellina, 2011; Ganly and Montgomery, 2012). These organizations react to changing market requirements with customized software or “skunk works” (Goodhue et al., 2009), the latter typically referring to manual spreadsheet-based tools (Prouty and Castellina, 2011). Surveys show that opinions are split on whether sticking to an ERP system is beneficial or not; some practitioners argue that ERP systems support the adaptation to changing conditions (Goodhue et al., 2009), while others argue that the inflexibility of ERP systems is a fundamental problem (Fauscette, 2013; IDG Market Pulse, 2013). Case evidence from dynamic environments is split, too, with some findings favoring ERP systems (Sarker et al., 2012) and others supporting standalone tools (Upton and Staats, 2008; Shaikh et al.,

149

2011). Yet, empirical tests comparing the effectiveness of the two approaches have not been published. The third capacity of dynamic capabilities, transforming, draws attention to the fact that adapting to a change in the market requirements is never a one-off undertaking but leads to further adjustments, as the organization learns more about the new requirements (Teece, 2007). This is where the literature is most critical of ERP systems. Even if one would accept the idea that ERP systems may initially facilitate the redesigning of processes (i.e., the seizing capacity), it is more difficult to fathom their support for continuous small adjustments (i.e., the transforming capacity). This is because the formal procedures supporting the former tend to be incompatible with the kind of improvisation that is needed for the latter (Pavlou and El Sawy, 2010). Since ERP systems control how processes are executed, all changes must first be configured to the rules of the system before the changed processes can be deployed (Dechow and Mouritsen, 2005; Lindley et al., 2008). This runs counter to the logic of evolutionary improvement, where the best processes emerge from constant experimentation (Burgelman, 1991). Furthermore, ensuring that all changes are adequately taken into account in all parts of the integrated enterprise-wide system necessitates centralized governance of the reconfiguration work, which tends to slow the process reengineering (Tiwana and Konsynski, 2010). While this may be acceptable for the initial process design, it is likely to frustrate initiatives for smaller improvements (Pavlou and El Sawy, 2010), thus hindering organizations’ overall ability to adapt to changing market requirements. In sum, the literature suggests that organizations facing dynamic market requirements may benefit from ERP systems’ information-processing capabilities, as they support the capacity to sense changes. However, this benefit may not be enough, as the literature is either mixed or outright negative about ERP systems’ effect on organizations’ ability to adapt their processes to observed changes. The net effect on operational performance remains to be tested. 3. Competing hypotheses Now that we have concluded that the results and arguments presented in the existing literature are mixed about the performance effects of using an ERP system under dynamic market requirements, we will next demonstrate that the question cannot be solved based on theoretical reasoning either. That is because the arguments for both the negative and the positive effects can be backed by different theories on organizational adaptability. 3.1. The view against ERP systems The operating logic of ERP systems is that every process is first defined in the configuration of the software and then executed exactly as defined, as ensured by the various checks and safeguards of the system (Kallinikos, 2004; Lowe and Locke, 2008). Through their emphasis on (i) process standardization, (ii) rulesbased control, and (iii) intensive collection of data into a centralized database, ERP systems represent one of the most profound contemporary manifestations of the Bureaucratic Theory of organizations (Kallinikos, 2004; Lowe and Locke, 2008; Morton and Hu, 2008; Berente and Yoo, 2012; Christiansen et al., 2012). This classic theory, rooted in the ideas of Weber (1946), Taylor (1911), and Fayol (1949), holds that standard processes, rules, and careful documentation of organizational activities are vital for the successful management of complex organizations where individuals and subunits are highly specialized (e.g., Blau, 1970). The specialization is necessary because no individual can master every task in a modern

150


industrial organization like they could in a traditional craftsman’s shop (e.g., Child, 1973), but it may lead to coordination issues because individuals who specialize in different tasks tend to perceive their objectives differently (e.g., Lawrence and Lorsch, 1967). Bureaucratic Theory offers the standard processes, rules, and the intensive centralized information collection as a means to control behavior so that the specialization would not distract individuals from pursuing the overall objectives of the organization (e.g., Weber, 1946). Despite its negative connotation, the bureaucratic system has persisted in many contemporary settings (Walton, 2005; Bunderson and Boumgarden, 2010). In fact, people have seen parallels between the core operations management philosophies and the principles of Bureaucratic Theory (Boje and Winsor, 1993), which is not surprising considering that bureaucracy’s objective to reduce process variation has also been shared by much of the operations management literature (e.g., Schmenner, 2012). The benefits of reduced process variation have been particularly emphasized in the enterprise software research (e.g., Gattiker and Goodhue, 2005; Cotteleer and Bendoly, 2006; Cotteleer, 2006; Seddon et al., 2010; Mithas et al., 2011). However, when an organization faces dynamic market requirements, the problem with the bureaucratic system is that it may take considerable time to change the standard processes, the rules that enforce them, and the data structures that support them. This idea is apparent in the previously reviewed literature on ERP systems’ shortcomings in seizing emerging opportunities and transforming processes to better exploit them. It is also evident in the comments of managers who have famously lamented that “installing ERP systems is like pouring concrete on a firm’s business processes” (Gattiker et al., 2005, p. 88). Despite the software vendors’ efforts to make ERP systems more easily adjustable, complaints about their inflexibility have persisted (Henschen, 2011; Ganly and Montgomery, 2012; IDG Market Pulse, 2013), suggesting that the systems’ perceived rigidity may be rooted deeper than in how the software is coded. Such an idea is supported by the Organic Theory of organization design, which has long warned against the use of the bureaucratic system in dynamic operating environments and advocated instead an “organic” approach based on informality and decentralization (Burns and Stalker, 1961). Similar advice has been prevalent in the operations management literature, where studies have found that more loosely defined and less tightly controlled processes are often best able to adapt to changing conditions (e.g., Nahm et al., 2003; Germain et al., 2008; Huang et al., 2010). The literature on Organic Theory argues that the bureaucratic system is fine in stable operating environments, where efficiency is the main priority, but ineffective or even detrimental in dynamic market conditions where adaptability is crucial and the organic approach is needed instead (Burns and Stalker, 1961). This contingency view on the context-dependent applicability of the two management approaches, bureaucratic and organic, has been corroborated by empirical organization design research (see reviews in, e.g., Donaldson, 2001; Scott, 2002). In the domain of ERP systems, it remains to be tested, although some studies have already implied the idea. Wang et al. (2012) found that information technology (IT) resources, such as ERP systems, have direct positive performance effects only under stable market conditions, and although they did not specifically analyze the effect of ERP systems, they used ERP systems in the examples of why the dynamism of market conditions can be expected to negatively moderate the performance effect of IT resources. Lu and Ramamurthy (2011) found that IT spending can be detrimental to an organization’s market capitalizing agility, and although they did not focus on ERP system spending, their discussion of the effect was largely based on the ERP systems literature. The contingency view was most explicitly articulated by Barki and Pinsonneault (2005) who proposed, based on a conceptual

analysis, that ERP systems should have negative performance effects in dynamic operating conditions. We formalize this view on ERP systems as follows: H1: The interaction between the use of an ERP system and the dynamism of an organization’s market requirements is negatively associated with operational performance. That is, extensive use of an ERP system under dynamic market requirements is negatively associated with operational performance. 3.2. The view in favor of ERP systems The negative view can be contested by proposing that the rigidities of ERP systems are actually a blessing in disguise. Such an argument can be based on Rigid Flexibility Theory, which challenges the tenets of Organic Theory by claiming that “an atmosphere of permissiveness cannot be tolerated” in dynamic operating environments (Collins and Schmenner, 1993, p. 444). Contrary to Organic Theory’s advice to accept the complexity of the front-line work and permit “continual redefinition” of organizational processes in dynamic environments (Burns and Stalker, 1961, p. 121), Rigid Flexibility Theory proposes simplicity and discipline as the solutions for coping with dynamic conditions (Collins and Schmenner, 1993). The concept of simplicity is drawn from the principles of the Toyota Production System, which state that there should always be one clearly specified way to carry out every process (Spear and Bowen, 1999). The discipline, in turn, refers to the organizational arrangements aimed at ensuring that the one specified way is consistently adhered to, including the explicit definition of an organization’s objectives, careful documentation of processes, and systematic measurement of performance (Collins et al., 1998). When processes are based on simplicity and discipline, the argument goes, they are “easier to reconfigure and adapt to changing [market] requirements” (da Silveira, 2006, p. 933). Several different mechanisms can be proposed to contribute to this effect in the context of ERP systems. Although “simplicity” is not the word most commonly associated with ERP systems, many have argued that when market requirements change, it is simpler to make the necessary process changes to one integrated software package than to several standalone tools and worry if they will function together afterwards (Gattiker et al., 2005; Seddon et al., 2010; Michael et al., 2012). Moreover, ERP systems’ built-in process templates may simplify the task of figuring out how the processes should be changed (Goodhue et al., 2009; Scott, 2011). And even if a fitting template were not immediately available, it may soon be added because the vendors of ERP systems keep track of their customers’ operating environments and collaborate with them to solve emerging challenges (Goodhue et al., 2009; Sarker et al., 2012). These reasons may explain da Silveira’s (2006) result on ERP systems contributing to the simplicity of process management. The use of an ERP system has also been seen as a contributor to the discipline dimension of Rigid Flexibility Theory (Snider et al., 2009). At least ERP systems should contribute to the three aforementioned constituents of discipline described by Collins et al. (1998). That is, when an ERP system is used, a clear definition of organizational objectives is a prerequisite (Yeh and Xu, 2013), the documentation of process steps is necessary (Dalal et al., 2004), and the systematic measurement of performance is one of the most praised outcomes (Seddon et al., 2010). Of additional importance is the fact that discipline is manifested in how processes are changed, as ERP systems provide formal procedures to ensure that every change to the software’s configuration is compatible with all features and rules that are used elsewhere in the organization (Gattiker et al., 2005; Scott, 2011). The meticulousness of these procedures may slow down the reactions to the changes in market


requirements and may thus be perceived as inflexibility (Lindley et al., 2008; Pavlou and El Sawy, 2010). However, the reward may be that in the end, the changed processes are more likely to function as expected and not cause conflicts between different activities. The question whether the net effect of subjecting an organization to this type of discipline is positive or negative has yet to be tested. Hints toward a possibly positive effect can be found outside the ERP system research. First, the built-in formal procedures to reconfigure ERP systems may constitute metaroutines that embody the best practices of process reengineering, akin to the process improvement procedures of quality management programs (e.g., Schroeder et al., 2008). Second, even if this were not the case and the formal procedures did not ensure the effectiveness of the redesigned processes, their existence may reduce ambiguity about managerial responsibilities, speeding up decision making and the reorganization of resources (Sine et al., 2006; Patel, 2011). Third, the formal procedures may relieve the stress and mitigate the cognitive limitations of those in charge of redesigning the processes (Juillerat, 2010). Lastly, research has shown that the key aspects of the bureaucratic system have also other benefits for organizations that face dynamic market requirements. More specifically, standardization and rules may help organizations maintain functional integration during the reengineering of processes (Swink and Nair, 2007), while the standardization together with the centralized repositories of process information make organizations more robust by reducing dependence on individual employees (Briscoe, 2007). In conclusion, it is also possible to argue the exact opposite of Hypothesis 1 and propose the following: H2: The interaction between the use of an ERP system and the dynamism of an organization’s market requirements is positively associated with operational performance. That is, extensive use of an ERP system under dynamic market requirements is positively associated with operational performance. 4. Empirical context 4.1. Manufacturing planning and control The two competing hypotheses are tested in the context of manufacturing planning and control. We chose this context mainly because, both historically and technically, it belongs to the very core of modern ERP systems (Jacobs and Weston, 2007). Hence, a module for manufacturing planning and control is available in all contemporary ERP software packages, and it has also been installed in most manufacturing firms (Olhager and Selldin, 2003; Harris and Davenport, 2006). Therefore, variance in the use of the manufacturing planning and control features should not be biased by the firm-specific choices of software or the special needs of different industry sectors, which would influence the use of many other ERP system modules, such as product configurators or project management tools. Equally importantly, the context was chosen because variance is known to exist in the use of ERP system-based manufacturing planning and control features—not every organization actually uses the features even if they have been installed (Tenhiälä, 2011). This is because standalone software for manufacturing planning and control activities is widely available (e.g., Brandl, 2011; ARC, 2013), because the use of in-house developed software is common (e.g., Shaikh et al., 2011; Carvalho et al., 2014), and because many of the activities can be performed even in spreadsheets (e.g., Prouty and Castellina, 2011; Berente and Yoo, 2012). Virtually all major ERP software vendors have structured their manufacturing planning and control modules so that the key activities are packaged into sub-modules that distinguish between the activities for planning and control and between the activities related to materials and capacity. The labels of the sub-modules

151

vary by vendor, but in general, they comprise (i) materials planning, (ii) capacity planning, (iii) materials management, and (iv) shop-floor control (e.g., SAP, 2013; Oracle, 2013; Microsoft, 2013). This arrangement prevails because it originates from the core functions of manufacturing resource planning (MRPII) systems, from which the ERP systems have evolved (Jacobs and Weston, 2007). The arrangement differs slightly from the textbook presentations of MRPII (e.g., Jacobs et al., 2011) in the sense that aggregate and detailed planning levels are typically not clearly separated. This is because modern computers no longer have the computational limitations to necessitate the separation and because the planning activities at both levels rely mostly on the same data. Fig. 1 illustrates the activities and their input information flows from the other business functions of a firm. 4.2. Contextualization of the constructs Because the features for the different manufacturing planning and control activities are typically packaged in separate submodules of an ERP system, it is possible to use only one or some of the features and not others. Therefore, we study the use of an ERP system in terms of the extent of its use, which is a variable that ranges from none to all four of the activities in Fig. 1. Considering that the controversy over ERP systems’ effects in dynamic market conditions has centered on whether they facilitate or impede organizations’ efforts to adapt processes to changing requirements (i.e., the seizing and transforming capacities), the moderator of our hypotheses must capture changes at such level of customer needs that they necessitate redesigning of production processes. This means that the variable cannot be based on changes in the mere timing or volume of demand, or in how the demand is distributed across the existing product variety. (The capability of handling this kind of variability is of interest too, but because organizations should be able to accommodate it with their existing processes, this capability will be considered as one dimension of operational performance, namely manufacturing flexibility.) To necessitate redesigning of existing processes, the customerinduced changes must be such that they influence the product variety that the organization offers. So, for the purposes of this study, we define the dynamism of market requirements as the rate of such changes in customer needs and preferences that make existing products obsolete and new product introductions necessary. This type of dynamism poses a challenge to manufacturing planning and control regardless of whether an ERP system or other software is used. When a manufacturing process for a new product is designed, one must describe for the software how the process is handled in the planning and control activities and also how the process relates to the other business functions (Fig. 1), if those connections are managed in the software. The latter aspect makes process changes so laborious in ERP systems. Due to the crossfunctional connections, new processes must be defined to the finest detail before they can be deployed in an organization that uses an ERP system (this being the seizing challenge discussed earlier). The cross-functional connections create another challenge once the process has been deployed and people learn that something should be changed in it. For example, adding or dropping process steps or changing the resources that are used in them has implications across the system. Thus, to ensure the integrity of, say, the cost accounting data, all changes must follow a formal procedure (this being the transforming challenge). The choice not to use an ERP system gives an organization more freedom to decide how to manage the cross-functional connections. It may ease the deployment and adjustment of processes, but it may also lead to conflicts between activities. The more of the manufacturing planning and control activities are managed in an ERP system, the more there are crossfunctional connections to be handled within the system. Therefore,

152


Fig. 1. Manufacturing planning and control activities.

both the challenges and the advantages should be proportionate to the extent of ERP system’s use. Lastly, the dependent variable must reflect capabilities that are critically influenced by the activities of the chosen empirical context. From the commonly used dimensions of operational performance, speed, cost efficiency, manufacturing flexibility, and delivery reliability depend greatly on manufacturing planning and control. Thus, they are all used in this study. Meanwhile, quality, for instance, has more to do with how products are designed, what raw materials are used, and what is actually done on the shop floor rather than with how the procurement and production activities are planned and how the stock transfers and shop orders are processed.

We investigated the possibility of a non-response bias in three ways. First, we compared the average plant sizes (number of employees) and annual revenues between the respondents and the non-respondents. t-Tests revealed no statistically significant differences (p > .05). Second, we compared the averages of all studied variables between the early and the late respondents (i.e., the first and the last 25% of the respondents, Armstrong and Overton, 1977). Again, no statistically significant differences were found. Third, we tested for industry sector-based self-selection by comparing the industry classifications of the respondents and the non-respondents. Despite small variations in the response rates (Table 1), a chi-square test indicates no statistically significant differences in the representation of different sectors (2 = 12.45, p = .33).

5. Research design 5.2. Measures 5.1. Data We test the competing hypotheses with data from a survey conducted in 2008 in a sample of manufacturing plants in Finland. We first acquired a random list of 991 plants in the 12 largest manufacturing industry sectors from the national tax administration office. We contacted the responsible manufacturing managers both by mail and e-mail and gave them the option of responding to our questionnaire either on paper or on a website. We subsequently sent reminder e-mails after two weeks to managers who had not yet responded. We received 151 usable responses, for a response rate of 15.2 percent. Table 1 provides an overview of the sample.

The method of strong inference is recommended for situations where existing theories lead to two equally justified yet contradictory propositions (Platt, 1964). This method is only applicable when the two hypotheses are mutually exclusive and neither is favored by the measurement or the testing method. In this study, both criteria are satisfied because the key independent variable, the extent of ERP system’s use, constitutes a continuum where low values represent lack of use and high values represent extensive use. Therefore, support for one hypothesis automatically leads to the rejection of the other and a possible failure to find any significant interaction effect leads to the rejection of both hypotheses. In the questionnaire, we measured the independent variable by


153

Table 1 Sample statistics. Industry sector

ISIC

Food products and beverages Wood and wood products, except furniture Paper and paper products Chemicals and chemical products Rubber and plastics products Other non-metallic mineral products Primary metals Fabricated metal products, except machinery Machinery and equipment n.e.c. Electrical machinery and apparatus n.e.c. Radio, television, and communication equipment Motor vehicles, trailers, and semi-trailers

15 20 21 24 25 26 27 28 29 31 32 34

ISIC: International Standard Industrial Classification

Total:

asking what system was used in the respondent’s plant to conduct the four activities depicted in Fig. 1. In addition to the ERP system, the options included spreadsheet software, in-house developed software, manual handling, and other respondent-specified systems. We also asked for the name of the ERP system to ensure that no one had misidentified some other software as an ERP system. This part of the questionnaire was presented as a matrix where the above options constituted the columns and the four manufacturing planning and control activities were shown on the rows. All respondents found their choices among the given options, and no one left any of the questions unanswered. We constructed the variable from the sum of the referrals to the plant’s ERP system, resulting in an index that ranged from zero to four. For the dynamism of market requirements, we used a scale validated by Kristal et al. (2010). For the performance dimensions of speed, cost efficiency, and manufacturing flexibility, we reviewed a variety of different measures (e.g., Rosenzweig et al., 2003; Swink et al., 2005; Schmenner and Vastag, 2006; Swink and Nair, 2007; Kristal et al., 2010) and chose the items that made the most sense

Responses

Sample size

Response rate

12 12 4 7 12 17 7 27 31 14 2 6

101 76 40 74 71 100 38 177 193 57 38 26

12% 16% 10% 9% 17% 17% 18% 15% 16% 25% 5% 23%

151

991

15%

in the context of this study and had exhibited good psychometric properties in prior work. The resulting scales are composed of questions that characterize the performance of the respondent’s plant relative to competitors. This approach was used because numerical measures of the aforementioned performance dimensions are seldom available and hardly commensurable among different plants. The validity and reliability of all four reflective scales were tested in a confirmatory factor analysis using Mplus 7 software. Table 2 shows the results of maximum likelihood estimation with robust standard errors (Yuan and Bentler, 2000). The measurement model is satisfactory. First, convergent validity is supported, as the items load significantly on their hypothesized factors, and all standardized loadings are reasonable. Second, Fornell and Larcker’s (1981) condition for discriminant validity is satisfied because the average variance extracted proportions range from .50 to .63 and are thus considerably greater than the squared correlations between the variables (Table 3). Third, the composite reliability indices range from .75 to .83, indicating no problems with measurement reliability. Fourth, the overall fit of the model is acceptable, as the chi-square statistic is not significant

Table 2 Validity and reliability analysis of the reflective measurement scales. Standardized item loadings “How does your plant perform in comparison to your competitors in. . .” (1: far worse; 4: about same; 7: far better) Speed (CR = .83, AVE = .62) “. . .order fulfillment speed?” .89** “. . .delivery speed?” .75** “. . .manufacturing throughput time?” .71** Cost efficiency (CR = .77, AVE = .54) .72** “. . .direct manufacturing costs?” “. . .total product costs?” .84** “. . .raw material costs?” .62** Manufacturing flexibility (CR = .75, AVE = .50) .66** “. . .delivery flexibility?” “. . .flexibility to change production volume?” .72** “. . .flexibility to change product mix?” .73** “Indicate the rate of change for the following issues in your operating environment:” (1: very slow; 5: very rapid) Dynamism of market requirements (CR = .83, AVE = .63) .80** “the rate at which your products and services become outdated” “the rate of new product and service introductions in your markets” .86** “the rate of change in customer needs and preferences in your industry” .71** Latent variable correlations .15 Speed – cost efficiency .27* Speed – flexibility Speed – dynamism .18† Cost efficiency – flexibility −.17 −.03 Cost efficiency – dynamism .21† Flexibility – dynamism 2 = 62.62, degrees of freedom (d.f.) = 48, p = .077, 2 /d.f. = 1.30. Comparative fit index (CFI) = .972, non-normed fit index (NNFI) = .961. Root mean square error of approximation (RMSEA) = .045 [90% confidence interval: .000–.074]. Standardized root mean square residual (SRMR) = .052. CR: composite reliability; AVE: average variance extracted. * p < .05. ** p < .01. † p < .10.

154


Table 3 Descriptive statistics and correlations.

1 2 3 4 5 6 7 8 9 10 11 * ** † a b

Variable

Mean

1

2

3

4

5

6

7

8

9

10

Product varietya Just-in-time production methods Age of the current system (months)a Dynamism of market requirements ERP system’s use Speed Cost efficiency Manufacturing flexibility Delivery reliability On-time delivery (%) Lateness (days)

54.53b 2.49 55.01b 2.61 2.54 4.76 4.11 5.06 5.02 91.70 4.53

196.70b 1.02 49.48b .75 1.40 1.05 .73 .58 1.18 12.14 4.73

.08 .16† .05 .04 .01 −.12 .15† .00 −.12 −.05

−.04 .15† .20* .15† .20* .14† .06 −.06 −.01

.09 .30** .02 .01 −.02 .08 −.01 .20*

.11 .20* −.04 .25** .01 .06 −.07

.07 .03 .09 .07 −.09 .16†

.17* .32** .49** .24** −.25**

−.21** .13 .05 −.15

.08 .01 .09

.43** −.14†

−.36**

p < .05. p < .01. p < .10. Logarithmic transformation is used in the analyses to remedy positive skewness. Untransformed values are shown for clarity.

and the other fit indices pass all commonly used thresholds (e.g., Hair et al., 2005). In contrast to speed, cost efficiency, and manufacturing flexibility, there is one dimension of operational performance, namely delivery reliability, which can be measured with commensurable numerical scales. We measured this dimension by asking the respondents about their plants’ on-time delivery performance as (i) the percentage of timely deliveries among all deliveries and (ii) the average number of days that late deliveries are delayed. The benefit of using numerical scales is that they can be argued to increase the objectivity of measurement even if they are self-reported (Rosenzweig and Roth, 2004). Yet, to be coherent with the other performance measures, we also included a question similar to the ones in Table 2 to capture the perceived delivery reliability relative to the competition. The three items were not aggregated into a single factor due to the disparity among the metrics and because the results are more informative when the effects on each item are analyzed separately. 5.3. Addressing common method bias and criterion validity concerns In the operations management research community, the survey method of data collection has faced criticism that cannot be overlooked (e.g., Singhal, 2008). The main critique stems from concerns about common method bias (CMB) and criterion validity. In this study, the survey method was chosen because it enabled us to obtain cross-sectional data on such intimate plant-level phenomenon as which information systems are used in specific activities. However, the two concerns remain, and thus their potential effects must be investigated separately. Although CMB can be a problem in any dataset, surveys are particularly vulnerable to it due to certain human tendencies (e.g., disposition for socially desirable answers and susceptibility to halo effects) that may cause data to differ systematically from the reality that the respondents are expected to describe. There are various ways to address this issue in the survey design (e.g., ensuring anonymity and separating related questions within the questionnaire, e.g., Dillman et al., 2009) and by analyzing the data with methods such as Harman’s single-factor test (Podsakoff and Organ, 1986). These methods are used in this study, but they cannot guarantee the absence of CMB. We did not implement other methods because all of them have their own disadvantages while none of them can conclusively eliminate CMB (Podsakoff et al., 2003). A more important reason, however, was that the hypotheses are tested in an interaction analysis, which eliminates the risk of CMB leading to false positive results (i.e., a type I error). This is because

CMB can only deflate the coefficients of interaction terms, and therefore, any possible CMB in the data only makes the analyses more conservative (Evans, 1985; Siemsen et al., 2010). The only risk is not getting support for either hypothesis even if one of them is actually true (i.e., a type II error). The concern about the criterion validity of survey data applies especially to performance measures (Singhal, 2008). Again, the survey method offered benefits (e.g., the ability to measure performance relative to competition, which in cross-sectional studies may produce more valid variables than numerical performance metrics, e.g., Ketokivi and Schroeder, 2004), but the concern remains over whether the respondents reveal true performance. Again, the potential problem is more about type II errors than the more severe type I errors; nevertheless, testing the criterion validity with objective secondary data is important for the credibility of the data. A problem lies in the fact that it is nearly impossible to collect cross-sectional objective data on operational performance at the level of individual manufacturing plants. Consequently, we test the criterion validity based on an assumption that the plants’ operational performance is positively correlated with firm profitability. Naturally, this approach has some challenges as well. First, profitability is blurred by the performance of business functions other than manufacturing, and thus a perfect correlation should not be expected. Second, most plants represent only single parts of firms’ plant networks, and in such situations, the individual plant’s contribution to the firm’s profitability is diluted significantly. To use profitability in the criterion validity test, we identified those plants that constituted firms of their own and used that subsample (n = 44) to analyze the correlations between the perceived measures of operational performance and the return on sales statistics that we acquired in 2009 from the government registry of financial statements. The results provide support for the criterion validity of the perceived measures; cost efficiency has the highest correlation of .50 (p < .001), followed by speed with .41 (p < .01) and manufacturing flexibility with .32 (p < .05). Only the perceived delivery reliability has a statistically insignificant correlation of .10, but it too is in the expected direction. Overall, the correlations are equal to or higher than those reported for similar comparisons in other studies (Swink et al., 2007; Menor and Roth, 2008; Kristal et al., 2010; Terjesen et al., 2011). To test the criterion validity of the dynamism variable, we followed Rosenzweig (2009) and used industry statistics as a proxy. From the available statistics, the best one for this purpose describes the proportion of revenues from new products during the studied year (Statistics Finland, 2010). This measure should be correlated with the industry sectors’ averages of the perceived


measure of the rate of change in customer preferences. Indeed, the correlation is .66 (p < .05), lending support to the criterion validity of the dynamism variable. A perfect correlation could not be expected because the statistics are only published at the level of two-digit industry codes and not all operating environments are identical at such high level of aggregation. 5.4. Control variables In addition to the variables of theoretical interest, we collected data on typical control variables, such as plant size and revenue, as well as on various non-software-based manufacturing planning and control practices. From the former, only the number of different products had a significant effect on any of the dependent variables, and it is therefore included in the reported analyses. The non-software-based manufacturing practices (e.g., kanban systems, priority sequencing rules, and line-flow layouts) turned out to be intercorrelated, which is understandable due to their combined use in just-in-time (JIT) production (Shah and Ward, 2007). To avoid multicollinearity while controlling for the effects of these practices, we had to create a summary variable that captures the effects of all individual items. This was achieved with the average of two broad Likert-scale questions about the “use of JIT production methods” and the “investment of resources, time, and/or money in JIT production methods over the last two years” (anchors = 1: no use/no investment and 5: extensive use/substantial investment). As for the software, possible learning effects are taken into account with a question about the age of the current manufacturing planning and control system (months). Lastly, dummy variables are used to control for industry sector-based differences in performance. 6. Results 6.1. Hypothesis testing Table 3 shows the descriptive statistics and the intercorrelation matrix for the variables used in the analyses. Factor scores were used to measure the variables with reflective measurement scales (i.e., speed, cost efficiency, manufacturing flexibility, and the dynamism of market requirements). Hierarchical regression analyses were used to test the hypotheses. The control variables were entered in the first step; the main-effect variables, the dynamism of market requirements and the extent of ERP system’s use in the second step; and their meancentered multiplicative interaction term in the third step. Table 4 shows the results on the performance variables with the reflective measurement scales while Table 5 shows the results on the three indicators of delivery reliability. Since both of the numerical measures of delivery reliability are censored (timeliness at 100 percent and lateness at zero days), ordinary least squares estimation would have been biased (Long, 1997). Therefore, we used maximum likelihood estimation with robust standard errors (Yuan and Bentler, 2000), embedded in the Mplus 7 software. To be consistent across all dependent variables, we used the same estimator in all of the regression analyses. The results support Hypothesis 2 about ERP system’s use being beneficial in organizations that face dynamic market requirements. The interaction effect between the dynamism of market requirements and the extent of ERP system’s use is positive and statistically significant on speed, manufacturing flexibility, and the perceptual measure of delivery reliability. It is negative and significant on the average lateness of late deliveries, which is a reversed measure of performance. On the percentage of timely deliveries, the interaction effect is positive but only approaching statistical significance. On cost efficiency, the interaction effect is positive but insignificant.

155

To better understand the lack of a significant effect on cost efficiency, we reran the analyses separately for each of the three items of the scale. A considerable discrepancy was found between the results on the direct manufacturing costs and raw material costs. For the former, the coefficient of the interaction term was .17 (p < .05), whereas it was effectively zero (p = .52) for the latter. This is understandable because although poor manufacturing planning and control may increase raw material costs (due to a need for larger inventory buffers and occasional expediting of inbound shipments), the relationship with actual production costs is much clearer. Effective manufacturing planning and control should directly reduce production costs by optimizing capacity utilization and eliminating problems such as rework and cannibalization. In contrast, raw material costs are also influenced by many other factors, including supplier selection and the quality of purchased materials. Perhaps these considerations have motivated some researchers to use narrower scales and measure cost efficiency solely through direct manufacturing costs (e.g., Cua et al., 2001; Peng et al., 2008; Hallgren and Olhager, 2009). Fig. 2 visualizes the results by plotting the conditional performance effects of the extent of ERP system’s use across the empirically valid range of the dynamism of market requirements (1.28–4.55). The slopes show how the coefficients of ERP system’s use depend on the dynamism variable. The shaded areas show 95% confidence bands, computed with the tools of Miller et al. (2013). This method of probing interaction effects has two advantages over the more common “pick-a-point” technique, popularized by Aiken and West (1991). First, it emphasizes an often-overlooked fact that the effects are seldom significant across the entire range of the moderator. In this study, only the positive effects reach significance, and without the confidence bands, one could erroneously conclude that ERP systems would be detrimental under non-dynamic market requirements. Second, comparing the points at which the effects become significant (i.e., the regions of significance, Preacher et al., 2006) gives information about the relative sensitivity of the different performance dimensions to the studied interaction. From the four performance dimensions that are measured in the same way (i.e., in relation to competition), the effects that become significant the earliest are those on manufacturing flexibility and delivery reliability, both at the dynamism level of 2.85, which corresponds to the 62nd percentile of the dynamism of market requirements in the sample. Next is the effect on speed, which becomes significant at the dynamism level of 3.24, corresponding to the 79th percentile. The effect on direct manufacturing costs becomes significant at the dynamism level of 3.30, corresponding to the 82nd percentile. In other words, when the dynamism of market requirements increases, the advantages of using an ERP system are most likely to show first in manufacturing flexibility and delivery reliability. The conditional effect plots also convey the same information as the “pick-a-point” technique because the effect sizes can be easily derived for any value of the moderator by multiplying the coefficient at the chosen value by the range of the independent variable. For example, at the dynamism level of 3.74 (i.e., 1.5 above the mean, which is a common choice to represent a high level of the moderator in the “pick-a-point” technique), the average estimated differences between the plants that do not use an ERP system in any of the studied activities and the plants that use it in all four activities are as follows: 1.0 points in the Likert scale of speed, .9 points in direct manufacturing costs, .5 points in manufacturing flexibility, and 1.5 points in delivery reliability. To put these figures into perspective, the average differences between non-users and heavy users of JIT production methods are .8 points in speed and .7 points in cost efficiency. In regard to the numerical measures of delivery reliability at the same dynamism level (3.74), the average differences between non-users and extensive users of ERP systems are

156

Table 4 Regression results on speed, cost efficiency, and manufacturing flexibility. Variable

Constant Food products and beverages Wood and wood products, except furniture

Chemicals and chemical products Rubber and plastics products Other non-metallic mineral products Primary metals Fabricated metal products, except machinery Electrical machinery and apparatus n.e.c. Radio, television, and communication equipment Motor vehicles, trailers, and semi-trailers Product variety (logarithm) Just-in-time production methods Age of the current system (logarithm)

Cost efficiency Step 2

Step 3

Step 1

Step 2

Step 3

Step 1

Step 2

Step 3

3.85** (.33) .57† (.33) .27 (.41) −.09 (.55) .44 (.45) .68* (.32) .71* (.30) .34 (.48) .09 (.28) .21 (.35) .82* (.35) .49 (.50) .01 (.07) .24** (.09) .04 (.05)

3.92** (.33) .43 (.32) .32 (.41) −.21 (.49) .44 (.44) .55† (.33) .85* (.33) .38 (.47) .16 (.28) .19 (.36) .72* (.31) .46 (.53) .00 (.07) .21* (.09) .04 (.05) .23* (.11) .04 (.07)

4.01** (.31) .46 (.32) .42 (.42) −.39 (.43) .37 (.42) .50 (.32) .69* (.33) .43 (.45) .08 (.28) .17 (.38) .73* (.29) .54 (.51) −.01 (.06) .19* (.09) .03 (.05) .21* (.10) .04 (.07) .18* (.07)

3.65** (.22) .27 (.18) .19 (.27) .29 (.50) .20 (.24) −.01 (.20) .48* (.22) 1.02** (.27) .38† (.19) .10 (.21) −.28 (.35) −.54† (.29) −.09† (.05) .18** (.06) .05 (.04)

3.65** (.23) .25 (.19) .19 (.27) .29 (.51) .19 (.24) −.02 (.21) .47† (.24) 1.03** (.27) .37* (.18) .09 (.21) −.29 (.35) −.55† (.30) −.09† (.05) .19** (.06) .05 (.04) .01 (.08) −.01 (.05)

3.68** (.23) .26 (.19) .22 (.27) .23 (.51) .17 (.24) −.04 (.21) .42† (.25) 1.05** (.27) .34* (.17) .09 (.21) −.29 (.36) −.52† (.30) −.09† (.05) .18** (.06) .05 (.04) .00 (.08) −.01 (.05) .06 (.05)

4.72** (.16) .05 (.17) −.22 (.27) .20 (.14) .26 (.20) .31† (.19) .00 (.15) −.20 (.14) −.10 (.16) .20 (.20) −.17 (.32) .44** (.17) .07* (.03) .06 (.04) −.02 (.03)

4.77** (.16) −.02 (.17) −.16 (.27) .13 (.14) .28 (.18) .24 (.17) .13 (.16) −.18 (.14) −.03 (.17) .22 (.21) −.22 (.28) .44** (.16) .07* (.03) .04 (.04) −.03 (.03) .14* (.07) .05 (.04)

4.81** (.15) −.01 (.17) −.12 (.26) .05 (.16) .25 (.17) .22 (.16) .06 (.16) −.16 (.15) −.07 (.17) .21 (.22) −.22 (.27) .48** (.16) .06† (.03) .03 (.04) −.03 (.03) .13* (.06) .05 (.04) .08* (.04)

.13 .13 1.42

.15 .02 1.78

.18 .02 4.61*

.21 .21 2.49**

.21 .00 .00

.21 .00 .99

.13 .13 1.37

.16 .03 2.90†

.18 .02 2.86†

Dynamism of market requirements ERP system’s use Interaction of dynamism and ERP system’s use R2 R2 F for R2

Manufacturing flexibility

Step 1

Unstandardized coefficients, standard errors in the parentheses. Machinery and equipment n.e.c. is the benchmark in the industry controls. * p < .05. ** p < .01. † p < .10.


Paper and paper products

Speed

Table 5 Regression results on delivery reliability. Variable

Constant Food products and beverages Wood and wood products, except furniture

Chemicals and chemical products Rubber and plastics products Other non-metallic mineral products Primary metals Fabricated metal products, except machinery Electrical machinery and apparatus n.e.c. Radio, television, and communication equipment Motor vehicles, trailers, and semi-trailers Product variety (logarithm) Just-in-time production methods Age of the current system (logarithm)

Lateness

Step 1

Step 2

Step 3

Step 1

Step 2

Step 3

Step 1

Step 2

Step 3

4.63** (.35) .28 (.40) .45 (.32) −.53† (.32) .80 (.53) .69† (.39) 1.02** (.36) −.22 (.43) −.39 (.32) .00 (.37) .25 (.36) −.40 (.38) −.01 (.07) .10 (.09) .14* (.06)

4.62** (.35) .39 (.42) .52† (.31) −.50 (.33) .87† (.52) .78† (.40) 1.14** (.37) −.25 (.44) −.31 (.32) .10 (.37) .30 (.39) −.29 (.37) −.01 (.07) .07 (.10) .12* (.06) −.06 (.13) .08 (.07)

4.76** (.33) .43 (.42) .65* (.33) −.77* (.36) .77† (.46) .71† (.38) .92* (.36) −.17 (.44) −.44 (.31) .07 (.36) .31 (.36) −.16 (.41) −.03 (.07) .04 (.09) .12* (.06) −.09 (.12) .08 (.07) .27** (.09)

87.9** (4.3) 15.4** (4.0) 10.9* (5.2) 13.0† (7.8) 15.8** (4.4) 13.5** (3.7) 17.6** (4.9) .6 (4.9) 6.9† (3.7) 5.3 (5.9) 10.3* (4.1) 6.7 (5.8) −.9 (1.1) −.5 (1.0) .8 (.8)

87.8** (4.2) 16.1** (3.8) 11.0* (5.3) 13.4† (7.4) 16.0** (4.5) 14.1** (3.5) 17.7** (5.2) .4 (5.0) 7.0† (3.8) 5.6 (5.3) 10.6** (3.9) 7.1 (5.4) −.9 (1.1) −.6 (1.0) .8 (.8) −.6 (1.4) .2 (.9)

88.7** (4.2) 16.5** (3.8) 11.9* (5.5) 11.4 (7.4) 15.4** (4.3) 13.5** (3.4) 16.3** (4.6) 1.0 (5.0) 6.1† (3.6) 5.4 (5.3) 10.7** (3.8) 8.0 (5.3) −1.0 (1.1) −.8 (1.0) .8 (.8) −.9 (1.4) .1 (.9) 1.9† (1.1)

7.53** (1.46) −6.78** (1.33) −.81 (2.30) −5.23** (1.14) −3.13 (2.25) −4.75** (1.30) −7.23** (1.68) −.09 (1.40) −2.79* (1.31) −3.07* (1.33) −3.35† (1.89) .83 (3.11) −.19 (.21) −.03 (.41) .37 (.23)

7.62** (1.47) −6.98** (1.40) −.74 (2.34) −5.55** (1.24) −3.13 (2.37) −4.94** (1.39) −7.07** (1.73) −.03 (1.43) −2.69* (1.36) −3.10* (1.35) −3.48† (1.95) .78 (2.97) −.19 (.22) −.06 (.41) .36 (.24) .32 (.63) .05 (.30)

7.16** (1.45) −7.10** (1.43) −1.16 (2.41) −4.70** (1.35) −2.75 (2.09) −4.69** (1.36) −6.38** (1.67) −.30 (1.43) −2.26† (1.34) −3.01* (1.38) −3.52† (1.84) .38 (2.75) −.14 (.22) .05 (.40) .37 (.24) .44 (.57) .07 (.29) −.88* (.39)

.17 .17 2.00*

.18 .01 .56

.23 .05 8.45**

.20 .20 2.40**

.20 .00 .08

.22 .02 3.35†

.28 .28 3.59**

.28 .00 .18

.31 .03 5.59*

Dynamism of market requirements ERP system’s use Interaction of dynamism and ERP system’s use R2 R2 F for R2

On-time percentage


Paper and paper products

Relative to competitors

Unstandardized coefficients, standard errors in the parentheses. Machinery and equipment n.e.c. is the benchmark in the industry controls. * p < .05. ** p < .01. † p < .10.

157

158


Fig. 2. Conditional effects of ERP system’s use.

9.2 percentage points in the timeliness and 3.7 days in the average lateness of late deliveries. 6.2. Post hoc analysis: does time moderate the effectiveness of an ERP system? The age of the current system for manufacturing planning and control turned out to influence only the perceived delivery

reliability measure. The lack of other effects, together with the overall support for the effectiveness of ERP systems, led us to wonder if the possible learning effect would only manifest itself as long as an ERP system is used. Such an effect could be observed as a three-way interaction among the dynamism of market requirements, ERP system’s use, and the age of the system. We investigated this possibility in a follow-up analysis and found some evidence for it. The three-way interaction effect on the average lateness of late


deliveries was −.46 (p < .05), explaining an additional 2.5 percent of the variance in it (F = 3.46, p < .05). This type of beneficial but modest time-based effect is in line with the earlier ERP system research (Bendoly and Schoenherr, 2005; Gattiker, 2007; Stratman, 2007). 6.3. Post hoc analysis: is an ERP system equally important in all activities? Our operationalization of the extent of ERP system’s use enabled us to investigate whether the effects of using the system varied between the different activities of manufacturing planning and control. This could be done by analyzing separately the effects of each of the four items that constituted the index for the extent of ERP system’s use. The results are shown in Table 6. All interaction effects are in the direction of Hypothesis 2, but some of them are insignificant, indicating that some differences indeed exist. The use of an ERP system in shop-floor control has a significant interaction effect on all but the cost dimension of performance. In capacity planning, the use of an ERP system has a significant interaction effect on four performance variables and one effect approaching significance. Using an ERP system in materials planning has three significant interaction effects and one that is approaching significance. In materials management, only one effect is significant and another one is approaching significance. 7. Implications 7.1. Practical implications This study was designed to address the practical question of whether the use of an ERP system improves or impedes performance of organizations that face dynamic market requirements. The result is clear: the use of an ERP system is beneficial, at least as far as manufacturing planning and control activities are concerned. It raises the question of how it is possible that so many practitioners, who complain about ERP systems’ inflexibility (Gattiker et al., 2005; Rettig, 2007; Goodhue et al., 2009; Ganly and Montgomery, 2012; IDG Market Pulse, 2013), have missed this positive effect. We think the reason is simply that the effect is not very obvious. As concluded in the literature review, an ERP system’s contribution to an organization’s capability to adapt its processes to changing market requirements can be equally justifiably argued to be positive or negative, or anything in between, depending on which aspect of the capability (i.e., the capacity for sensing, seizing, or transforming) is considered. Moreover, as evidenced by our discussion of the Organic Theory of organization design, one can reasonably suspect that ERP systems might be fundamentally unfit for all organizations that operate in dynamic conditions. In fact, ERP systems’ reputation for inflexibility is quite understandable considering many of the issues discussed in this paper. The process reengineering experience is doubtlessly frustrating when one must follow the system’s reconfiguration procedures, which are typically considered slow and onerous (Dechow and Mouritsen, 2005; Lindley et al., 2008), yet due to the many restrictions of the system, the outcome is likely to fall short from a perfect fit to the new requirements (Sia and Soh, 2007; Strong and Volkoff, 2010). When facing such experiences, it may be difficult to appreciate that the rigid procedures and restrictions of an ERP system may be there to help ensure that the changed processes will work without conflicts between interrelated activities. Similar systematic enterprise-wide compatibility checking is not enforced when changes are made to standalone software, let alone to spreadsheetbased “skunk works.” This may lead to a smoother experience for those who make the changes, but it is equally possible that it may lead to conflicts between activities that damage operational

159

performance more than the rigidities of an ERP system would have. This idea of a tradeoff between the ease of making changes and the rigidity of ensuring that they will be effective is at the root of Rigid Flexibility Theory (Collins and Schmenner, 1993), and it has also been proposed in two conceptual studies on ERP systems (Gattiker et al., 2005; Scott, 2011). However, the ongoing debate on whether ERP systems are beneficial or detrimental in dynamic conditions proves that the idea is anything but common wisdom. This is where the empirical evidence from this study can possibly make a difference and encourage managers to think twice if they consider dismissing or replacing their ERP system in some organizational activities on the grounds of its perceived inflexibility to process changes. 7.2. Theoretical generalizability When Platt (1964) originally described the strong inference research design of testing competing hypotheses, he meant it to be used to discriminate between correct and flawed theoretical explanations. However, when studying socio-technical systems, the possible points of view and variables of interest are so numerous that it would be impossible to design a study that would once and for all reject some previously supported theoretical perspective. Thus, the current use of strong inference in management research has been limited to test the applicability of alternative theoretical perspectives in specific situations (e.g., Das et al., 2008; Rungtusanatham et al., 2005; Shaw et al., 2005). That was the purpose of this study as well. Although the results of this study favor the hypothesis based on Rigid Flexibility Theory, it clearly does not mean that the Organic Theory of organization design must be somehow flawed. Considering the amount of empirical research supporting Organic Theory, such conclusion could not be drawn from any single study. Instead, there must be some contingency factor in our empirical context that makes Organic Theory less relevant in the studied situation. The possible contingency factors can be explored by identifying the boundary conditions of this study’s results (Dubin, 1978). They describe the relevant contextual factors that determine whether or when the results can be theoretically generalized to other contexts than the studied one. We can logically identify two boundary conditions that may limit the theoretical generalizability of our findings and explain why the results are in favor of using an ERP system when an organization faces changing market requirements. First, the manufacturing planning and control activities constitute a relatively technical system. Perhaps the alleged adverse effects of bureaucracy in dynamic conditions are not as severe in technical systems as in purely human systems. After all, coordination among people has been the focus of interest in the classic literature on Organic Theory (see, e.g., Donaldson, 2001), as well as in the related contemporary research, including studies into entrepreneurial activity (Goodale et al., 2011) and the management of “quality exploration” projects (Zhang et al., 2012). In manufacturing planning and control, coordination also involves materials, machines, and other nonhuman resources. Although human resources are important, the system is predominantly technical because the nonhuman resources are less flexible than people. For example, people can assume new tasks if necessary, but certain materials can be used only for specific products and certain machinery can only be used for specific operations. Despite component commonality and flexible manufacturing systems, the constraints are tighter than in systems where only human activities are coordinated. This may be the reason why in the studied context, the bureaucratic system responds to dynamic market requirements better than an organic one. The standard processes, rules, and the centralized information repositories of the bureaucratic system ensure that all technical constraints are taken into

160


Table 6 Activity-specific results of using an ERP system. Activity

Variable

Speed

Cost efficiencya

Manufacturing flexibility

Delivery reliability

On-time percentage

Lateness

Materials planning [62%]

Constant

3.80** (.35) .00 (.18) .24 (.20) .36† (.21)

3.57** (.37) −.28 (.21) .29 (.21) .27 (.27)

4.67** (.16) −.04 (.08) .18 (.11) .25* (.11)

4.47** (.35) −.67** (.20) .36† (.21) .86** (.24)

86.4** (4.2) −4.1* (2.1) 3.1 (2.5) 4.6 (3.1)

7.95** (1.66) 2.10* (.95) −1.04 (.78) −2.54* (1.22)

3.90** (.32) .17 (.13) .22 (.19) .20 (.19)

3.78** (.36) −.29 (.18) −.16 (.19) .50* (.25)

4.74** (.16) .10 (.07) .16 (.10) .11 (.12)

4.60** (.34) −.24 (.16) .34† (.18) .41† (.23)

89.0** (3.9) −2.6* (1.3) −6.1** (2.2) 4.3 (3.2)

7.18** (1.43) .74 (.69) 2.05* (.89) −1.00 (1.12)

3.97** (.36) −.05 (.20) .06 (.22) .41† (.21)

3.63** (.41) −.37 (.24) .21 (.24) .38 (.27)

4.80** (.19) −.03 (.09) .02 (.12) .22* (.11)

4.81** (.40) −.63** (.23) .01 (.24) .74** (.25)

84.7** (4.3) −5.1* (2.0) 5.3* (2.5) 5.6* (2.7)

8.00** (1.72) 2.09* (.96) −1.18 (.90) −2.37* (1.12)

4.19** (.34) −.07 (.19) −.23 (.18) .44* (.21)

3.77** (.38) −.31 (.25) .02 (.21) .31 (.28)

4.79** (.18) −.04 (.10) .02 (.11) .25* (.12)

4.75** (.37) −.53* (.22) −.01 (.21) .62* (.25)

86.9** (4.2) −4.9* (2.1) 2.3 (2.5) 5.5* (2.8)

7.38** (1.71) 2.06* (.88) −.30 (.87) −2.41* (1.07)

Dynamism ERP system Interaction

Materials management [42%]

Constant Dynamism ERP system Interaction

Capacity planning [77%]


Shop-floor control [72%]


Unstandardized coefficients, standard errors in the parentheses. Control variables are the same as in Tables 4 and 5, but here their coefficients are omitted for clarity. Percentage of plants using an ERP system in each activity is given in the square brackets. * p < .05. ** p < .01. † p < .10. a Cost efficiency column shows the effects on the direct manufacturing costs.

account and adhered to when an organization reacts to the changed market requirements. Another possible boundary condition is the interdependence of the studied activities. As shown in Fig. 1, the manufacturing planning and control activities are tightly connected, not only to one another but also to the other business functions. Thus, any changes in how the activities are performed can easily influence many other activities. If such interdependent activities are changed on an adhoc basis, the repercussions for the other activities and business functions can be severe. For example, a change in the materials planning activity could mean that its output can no longer be used in capacity planning; or a change in the shop-floor control could mean that shop orders are issued to the wrong personnel. The interdependence of activities may well favor ERP systems considering all the discussion in this paper about ERP systems’ reconfiguration procedures and some of their technical constraints possibly helping to ensure that the changed activities are compatible with the other activities performed in the organization. Such an idea would be aligned with the finding of Bendoly et al. (2008) that experienced managers view ERP systems as particularly useful under the conditions of tight task interdependence. The observations about the boundary conditions translate easily to propositions that can be tested in further studies as three-way interactions. For example, a measure can be developed for how human (as opposed to technical) an organizational system is or how tightly interdependent the studied activities are. Studying such

variables in future research would enable further theoretical development in the effects of ERP systems and further in the applicability and relative strengths of the Bureaucratic and the Organic Theories of organization design. 7.3. Implications for research on organizational adaptability Although Rigid Flexibility Theory was invented already more than two decades ago, it has not been used or tested in many empirical inquiries (notable exceptions include Collins et al., 1998; da Silveira, 2006; Snider et al., 2009). The sparse use is surprising considering the importance of understanding the different mechanisms that contribute to organizational adaptability. Perhaps the propositions of Organic Theory have become such common wisdom that there has been little demand for a theory that essentially questions the benefits of informality and decentralization. Only few earlier studies have casted doubt on the idea that Organic Theory would provide the universal solutions for dynamic environments (Sine et al., 2006; Patel, 2011). Based on the results of this study, we think that Rigid Flexibility Theory offers a promising perspective to complement our understanding of the different ways of developing organizational adaptability. One challenge in using Rigid Flexibility Theory is that the literature has been quite ambiguous about the theory’s key constructs, simplicity and discipline. While the original work of Collins and Schmenner (1993) equated both constructs with JIT production


practices, the subsequent studies have associated them with many other practices, too, including the use of an ERP system (da Silveira, 2006; Snider et al., 2009). Instead of equating the constructs with any specific practices, it could be beneficial for the further development of Rigid Flexibility Theory, if the constructs were defined in more general terms (yet precisely enough to ensure the “falsifiability” of the theory, as per Popper, 1959). Hence, we suggest that in Rigid Flexibility Theory, simplicity refers to the reduction of heterogeneity in the execution of organizational processes and in the management of the process-related information. Discipline, in turn, refers to the exercise of rules-based control over the execution of processes. These definitions are aligned with the ideas presented in this study as well as in the earlier investigations. Simplicity is manifested in the same heterogeneity-reducing manner in the standardized processes and centralized databases of ERP systems as it is in JIT production systems’ highly specified activities and yes-or-no type of communication (Spear and Bowen, 1999). Similarly, the definition of discipline as formal rules-based control is manifested equally well by the formal protocols that are embedded in ERP systems and the statistical process control techniques that have been discussed in the earlier work on Rigid Flexibility Theory (Collins et al., 1998). We also note that the proposed definition of simplicity is aligned with Miller and Friesen’s (1983) notion of heterogeneity constituting the essence of complexity. It is not aligned with the other classic view that considers complexity as a large number of parts interacting in nontrivial ways (e.g., Simon, 1962). Here, the heterogeneity-reduction perspective is more practical because it may be very difficult to reduce the number of parts or make their interactions completely predictable in real organizational settings. Meanwhile, it may be more feasible to reduce the different ways of executing the same activities in organizational processes or the different tools that are used to process the same information. Furthermore, the benefit of reducing the number of parts in a system or making their interactions more predictable could be considered almost self-evident, whereas the proposed benefit of reducing heterogeneity in organizational activities is much “riskier” in the Popperian sense, as it defies the evolutionary view of autonomous experimentation being crucial for organizational adaptability (Burgelman, 1991). Similarly, proposing that the rulesbased control supports organizational adaptability is anything but self-evident, as it defies the propositions of Organic Theory, as discussed earlier. Again, we do not suggest that the other perspectives to organizational adaptability are wrong; we simply propose that the rules-based control and the reduction of heterogeneity in organizational activities and information processing become significant contributors to organizational adaptability in the presence of either of the previously discussed boundary conditions, that is, in organizational systems that are predominantly technical in nature or where activities are tightly interdependent.

161

additional costs from, for example, manual coordination between activities that are managed with separate tools. Consequently, the effect on total costs may be either different from or similar to the observed effect on the direct manufacturing costs. The total cost effect could possibly be probed in further inquiries, but considering the difficulties of studying indirect costs, a case-study approach would probably be more effective than a cross-sectional research design. We also note that our measure of ERP system’s use has the shortcoming of not allowing the testing of its psychometric properties. The present operationalization is more appropriate than the more common reflective measurement mode in the sense that the use of a software tool for a given activity does not constitute a latent but a directly measurable construct (a.k.a. an operationally defined construct, Ketokivi and Schroeder, 2004), yet the inability to test the measure’s reliability may be a concern. At worst, some respondents may not have known what software tools are actually used in their plants. Although no one indicated this by leaving any of the questions unanswered, the correctness of the answers cannot be taken for granted. The ways to eliminate this concern are difficult to implement in a cross-sectional study because one should measure the use of the software through work observation or by collecting data directly from the organizations’ information systems. Such methods can be recommended if further inquiries are conducted using the case-study approach. Moreover, we note that the measure for ERP system’s use only captures whether it is used in the studied activities, not whether it is used exactly as it should be, nor whether the system has been implemented well. This is a limitation because both of these factors have been found to influence performance (e.g., Seddon et al., 2010). Therefore, our results represent the average effects across all organizations irrespective of the quality of use or the success of the implementation. The unaccounted-for variance in these two factors may have increased the standard errors of our coefficients, making the hypothesis testing more conservative and meaning that the actual effect in any individual organization may be greater or smaller than what the point estimates of the coefficients suggest. To obtain a more precise picture of the effect, measures of the two factors would have to be included as additional moderators in further research. Finally, the fact that the studied organizations’ need to make changes in their processes is measured through the dynamism of their market requirements may pose a limitation, too. Basically, it is possible that this type of dynamism leads to a specific kind of process changes that happen to be particularly well supported by ERP systems. It is therefore possible that the advantage of ERP systems would be diminished or even negated if the effects of some other types of dynamism were analyzed. The alternative types of dynamism could be related to changes in competition or in production technologies, for instance (e.g., Miller and Friesen, 1983). 8.2. Statistical generalizability

8. Limitations 8.1. Measurement issues The discussion of the onerous nature of ERP systems’ reconfiguration procedures hint that the use of an ERP system under dynamic market requirements may involve costs that are not captured in our measurement of cost efficiency. In terms of total costs, ERP systems’ positive effect on the direct manufacturing costs might well be undermined by the labor costs that stem from the reconfiguration work that is necessary when processes are changed. It is arguably less laborious to make adjustments to standalone software or especially to spreadsheet-based “skunk works” (e.g., Goodhue et al., 2009). On the other hand, those adjustments may also involve

The population of the studied sample is the 12 largest manufacturing sectors of a single country, and thus strict statistical generalizability of the results is limited. However, it may be possible to extrapolate the results to other countries where relevant industry characteristics are similar. One essential characteristic for such an extrapolation is the market penetration of ERP systems. In Finland, ERP systems are very common: 91 percent of manufacturers had implemented them already by 2006 (Aara, 2006). The situation is similar in many other European countries such as Sweden where the projected market penetration for the same time period was 89 percent (Olhager and Selldin, 2003). The figure is slightly lower in the U.S., where 74 percent of manufacturers had implemented ERP systems by 2010 (Jutras, 2010). This together

162


with the fact that we measured the main performance variables in relation to competition means that in the U.S., the potential advantage for any organization that makes extensive use of an ERP system can be greater than what the results of this study suggest. In developing countries, where the market penetration of ERP systems is even lower, the potential advantage should be even greater. Another relevant factor for generalizability lies in the structure of the manufacturing industry. In Finland, the largest industry sectors produce rather complex discrete products such as industrial machinery and instruments. It is likely that the advantages of ERP systems are emphasized in these sectors because the planning and control activities can be often simplified in process industries and in the repetitive manufacturing of simple discrete products (e.g., Jacobs et al., 2011). Thus, the average effects could be smaller in countries where the latter are better represented. A comparison of national economic statistics shows that, in Europe, the composition of the manufacturing industry in Austria, Denmark, Germany, Italy, and Sweden is very similar to that in Finland (Eurostat, 2013). In the U.S., the manufactures of food products, oil and gas, and chemicals are relatively more important, and thus, the average effect would probably be a little smaller. The same applies to most developing countries where also the manufactures of textiles, primary metals, and other mineral products are typically larger than the complex discrete manufacturing sectors. Nevertheless, there is no reason to expect differences in the effects if only the more complex discrete manufacturing sectors of these countries are considered. 9. Conclusions This study has provided an answer to the question of whether the use of an ERP system is beneficial or detrimental for organizations that face dynamic market requirements. The answer applies to manufacturing planning and control, and the results are strongly in favor of using an ERP system. The question was viewed from two theoretical perspectives in order to reach a more profound understanding of the phenomenon. However, the fact that the hypothesis based on the Organic Theory was rejected does not necessarily point to any flaws in that theory. Instead, there must have been something in the studied context that favored the ERP systems. The discussion about the boundary conditions thus serves as a steppingstone for further theoretical development. Toward this end, we suggested that the relevant boundary conditions might have been that the studied organizational system was predominantly technical (as opposed to human) and that the interdependence of the studied activities was very tight (as opposed to loose). We propose that if either of these contingency variables is included in a future study within some other context (where there is variance in them), then they may positively moderate the effectiveness of ERP systems under dynamic market requirements. With this study, we aim to make a contribution with substantive theory. In the terminology of Glaser and Strauss (1967), substantive theories are applications of formal theories in specific contexts. The results of this study constitute a substantive theory of the use of an ERP system in manufacturing planning and control. It is characteristic of substantive theories that they are of most interest only to a select group of people; however, through the exploration of relevant boundary conditions, wider implications can be drawn inductively. In this study, the above-summarized propositions together with our attempt to clarify the constructs of Rigid Flexibility Theory aim for a more general contribution. Future research on the proposed contingency variables and on the more generic operationalizations of the discipline and simplicity constructs may improve our understanding of the applicability of organic and bureaucratic solutions in dynamic environments. Such research is warranted, as the results of this study have shown that

the organic solutions may not be as superior as is often suggested in the literature.

Acknowledgements We are grateful for the feedback of the anonymous reviewers and the invaluable advice from the associate editor. We also thank Tom Gattiker, Jose´ Esteves, John Ettlie, Fabrizio Salvador, and Kari Tanskanen for their comments and suggestions. We further gratefully acknowledge financial support from Tekes—The Finnish Funding Agency for Technology and Innovation (Project SSOC) and the Spanish Ministry of Science and Innovation (Grant No. ECO2010-18293). Finally, we thank the informants from the studied organizations for making this research possible in the first place.

References Aara, 2006. ERP System Market in Finland 2005–2006. Aara Finland Oy, Helsinki, Finland. Aiken, L.S., West, S.G., 1991. Multiple Regression: Testing and Interpreting Interactions. Sage Publications, Newbury Park, CA. ARC, 2013. Manufacturing Execution Systems for Discrete Industries. ARC Advisory Group, Dedham, MA. Armstrong, J.S., Overton, T.S., 1977. Estimating nonresponse bias in mail surveys. J. Market. Res. 14 (3), 396–402. Barki, H., Pinsonneault, A., 2005. A model of organizational integration, implementation effort, and performance. Organ. Sci. 16 (2), 165–179. Bendoly, E., Bachrach, D.G., Powell, B., 2008. The role of operational interdependence and supervisory experience on management assessments of resource planning systems. Prod. Oper. Manage. 17 (1), 93–106. Bendoly, E., Cotteleer, M.J., 2008. Understanding behavioral sources of process variation following enterprise system deployment. J. Oper. Manage. 26 (1), 23–44. Bendoly, E., Rosenzweig, E.D., Stratman, J.K., 2009. The efficient use of enterprise information for strategic advantage: a data envelopment analysis. J. Oper. Manage. 27 (4), 310–323. Bendoly, E., Schoenherr, T., 2005. ERP system and implementation-process benefits: implications for B2B e-procurement. Int. J. Oper. Prod. Manage. 25 (4), 304–319. Berente, N., Yoo, Y., 2012. Institutional contradictions and loose coupling: postimplementation of NASA’s enterprise information system. Inform. Syst. Res. 23 (2), 376–396. Blau, P.M., 1970. A formal theory of differentiation in organizations. Am. Sociol. Rev. 35 (2), 210–218. Boje, D.M., Winsor, R.D., 1993. The resurrection of Taylorism: total quality management’s hidden agenda. J. Organ. Change Manage. 6 (4), 57–70. Boudreau, M.-C., Robey, D., 2005. Enacting integrated information technology: a human agency perspective. Organ. Sci. 16 (1), 3–18. Brandl, D., 2011. MES vs. ERP for manufacturing. Control Eng. 58 (11), 14–16. Briscoe, F., 2007. From iron cage to iron shield? How bureaucracy enables temporal flexibility for professional service workers. Organ. Sci. 18 (2), 297–314. Bunderson, J.S., Boumgarden, P., 2010. Structure and learning in self-managed teams: why “bureaucratic teams” can be better learners. Organ. Sci. 21 (3), 609–624. Burgelman, R.A., 1991. Intraorganizational ecology of strategy making and organizational adaptation: theory and field research. Organ. Sci. 2 (3), 239–262. Burns, T., Stalker, G.M., 1961. The Management of Innovation. Tavistock Publications, London, UK. Carvalho, A.N., Scavarda, L.F., Lustosa, L.J., 2014. Implementing finite capacity production scheduling: lessons from a practical case. Int. J. Prod. Res. 52 (4), 1215–1230. Collins, R.S., Cordon, C., Julien, D., 1998. An empirical test of the rigid flexibility model. J. Oper. Manage. 16 (2–3), 133–146. Collins, R.S., Schmenner, R., 1993. Achieving rigid flexibility: factory focus for the 1990s. Eur. Manage. J. 11 (4), 443–447. Cotteleer, M.J., 2006. An empirical study of operational performance parity following enterprise system deployment. Prod. Oper. Manage. 15 (1), 74–87. Cotteleer, M.J., Bendoly, E., 2006. Order lead-time improvement following enterprise information technology implementation: an empirical study. MIS Quart. 30 (3), 643–660. Cua, K.O., McKone, K.E., Schroeder, R.G., 2001. Relationships between implementation of TQM, JIT, and TPM and manufacturing performance. J. Oper. Manage. 19 (6), 675–694. Child, J., 1973. Parkinson’s progress: accounting for the number of specialists in organizations. Adm. Sci. Q. 18 (3), 328–348. Christiansen, U., Kjærgaard, A., Hartmann, R.K., 2012. Working in the shadows: understanding ERP usage as complex responsive processes of conversations in the daily practices of a Special Operations Force. Scand. J. Manage. 28 (2), 173–184.

A. Tenhiälä, P. Helkiö / Journal of Operations Management 36 (2015) 147–164 da Silveira, G.J.C., 2006. Effects of simplicity and discipline on operational flexibility: an empirical reexamination of the rigid flexibility model. J. Oper. Manage. 24 (6), 932–947. Dalal, N.P., Kamath, M., Kolarik, W.J., Sivaraman, E., 2004. Toward an integrated framework for modeling enterprise processes. Commun. ACM 47 (3), 83–87. Das, A., Pagell, M., Behm, M., Veltri, A., 2008. Toward a theory of the linkages between safety and quality. J. Oper. Manage. 26 (4), 521–535. Dechow, N., Mouritsen, J., 2005. Enterprise resource planning systems, management control and the quest for integration. Account. Organ. Soc. 30 (7–8), 691–733. Deloitte, 2010. Best-of-Breed Enterprise Applications. Deloitte Development LLC, Hermitage, TN. Dillman, D.A., Smyth, J.D., Christian, L.M., 2009. Internet, Mail, and Mixed-Mode Surveys: The Tailored Design Method. John Wiley & Sons, Hoboken, NJ. Donaldson, L., 2001. The Contingency Theory of Organizations. Sage Publications, Thousand Oaks, CA. Drobik, A., Rayner, N., 2011. How to Learn to Love Your ERP (Again). Gartner, Stamford, CT. Dubin, R., 1978. Theory Building, revised ed. Free Press, New York, NY. Eurostat, 2013. Structural Business Statistics (SBS), http://epp.eurostat.ec.europa. eu/portal/page/portal/european business/introduction (accessed 27.5.14). Evans, M.G., 1985. A Monte Carlo study of the effects of correlated method variance in moderated multiple regression analysis. Organ. Behav. Hum. Decis. Process. 36 (3), 305–323. Fauscette, M., 2013. Maintaining ERP Systems: The Cost of Change. IDC, Framingham, MA. Fayol, H., 1949. General and Industrial Management. Pitman Publishing, London, UK. Fornell, C., Larcker, D.F., 1981. Structural equation models with unobservable variables and measurement error: algebra and statistics. J. Market. Res. 18 (3), 382–388. Ganly, D., Montgomery, N., 2012. Hype Cycle for ERP 2012. Gartner, Stamford, CT. Gattiker, T.F., 2007. Enterprise resource planning (ERP) systems and the manufacturing-marketing interface: an information-processing theory view. Int. J. Prod. Res. 45 (13), 2895–2917. Gattiker, T.F., Chen, D., Goodhue, D.L., 2005. Agility through standardization: a CRM/ERP application. In: Bendoly, E., Jacobs, F.R. (Eds.), Strategic ERP Extension and Use. Stanford Business Books, Stanford, CA, pp. 87–96. Gattiker, T.F., Goodhue, D.L., 2005. What happens after ERP implementation: understanding the impact of interdependence and differentiation on plant-level outcomes. MIS Quart. 29 (3), 559–585. Germain, R., Claycomb, C., Dröge, C., 2008. Supply chain variability, organizational structure, and performance: the moderating effect of demand unpredictability. J. Oper. Manage. 26 (5), 557–570. Glaser, B.G., Strauss, A.L., 1967. The Discovery of Grounded Theory: Strategies for Qualitative Research. Wiedenfeld and Nicolson, London, UK. Goodale, J.C., Kuratko, D.F., Hornsby, J.S., Covin, J.G., 2011. Operations management and corporate entrepreneurship: the moderating effect of operations control on the antecedents of corporate entrepreneurial activity in relation to innovation performance. J. Oper. Manage. 29 (1–2), 116–127. Goodhue, D.L., Chen, D.Q., Boudreau, M.C., Davis, A., Cochran, J.D., 2009. Addressing business agility challenges with enterprise systems. MIS Quart. Exec. 8 (2), 73–87. Gupta, M., Kohli, A., 2006. Enterprise resource planning systems and its implications for operations function. Technovation 26 (5–6), 687–696. Hair, J.F., Black, B., Babin, B., Anderson, R.E., Tatham, R.L., 2005. Multivariate Data Analysis, 6th ed. Prentice Hall, Upper Saddle River, NJ. Hallgren, M., Olhager, J., 2009. Lean and agile manufacturing: external and internal drivers and performance outcomes. Int. J. Oper. Prod. Manage. 29 (10), 976–999. Harris, J.G., Davenport, T.H., 2006. New Growth from Enterprise Systems: Achieving High Performance through Distinctive Capabilities. Accenture Institute for High Performance Business, Wellesley, MA. Hendricks, K.B., Singhal, V.R., Stratman, J.K., 2007. The impact of enterprise systems on corporate performance: a study of ERP, SCM, and CRM system implementations. J. Oper. Manage. 25 (1), 65–82. Henschen, D., 2011. Enterprise Apps 2011: Focus on Business Relevance. InformationWeek Analytics, Manhasset, NY. Huang, X., Kristal, M.M., Schroeder, R.G., 2010. The impact of organizational structure on mass customization capability: a contingency view. Prod. Oper. Manage. 19 (5), 515–530. IDG Market Pulse, 2013. ERP Challenges and Successes. IDG Research Services, Framingham, MA. Jacobs, F.R., Berry, W.L., Whybark, D.C., Vollmann, T.E., 2011. Manufacturing Planning and Control for Supply Chain Management, 6th ed. McGraw-Hill, New York, NY. Jacobs, F.R., Weston Jr., F.C., 2007. Enterprise resource planning (ERP) – a brief history. J. Oper. Manage. 25 (2), 357–363. Juillerat, T.L., 2010. Friends, not foes? Work design and formalization in the modern work context. J. Organ. Behav. 31 (2–3), 216–239. Jutras, C., 2010. ERP in Manufacturing 2010: Measuring Business Benefit and Time to Value. Aberdeen Group, Boston, MA. Kallinikos, J., 2004. Deconstructing information packages: organizational and behavioural implications of ERP systems. Inform. Technol. People 17 (1), 8–30. Karimi, J., Somers, T.M., Bhattacherjee, A., 2007. The impact of ERP implementation on business process outcomes: a factor-based study. J. Manage. Inform. Syst. 24 (1), 101–134.

163

Ketokivi, M., Schroeder, R.G., 2004. Perceptual measures of performance: fact or fiction? J. Oper. Manage. 22 (3), 247–264. Kristal, M.M., Huang, X., Roth, A.V., 2010. The effect of an ambidextrous supply chain strategy on combinative competitive capabilities and business performance. J. Oper. Manage. 28 (5), 415–429. Lai, F., Li, X., Lai, V.S., 2013. Transaction-specific investments, relational norms, and ERP customer satisfaction: a mediation analysis. Decis. Sci. 44 (4), 679–711. Lawrence, P.R., Lorsch, J.W., 1967. Differentiation and integration in complex organizations. Adm. Sci. Q. 12 (1), 1–47. Lindley, J.T., Topping, S., Lindley, L.T., 2008. The hidden financial costs of ERP software. Manage. Financ. 34 (2), 78–90. Long, S.J., 1997. Regression Models for Categorical and Limited Dependent Variables. Sage Publications, Thousand Oaks, CA. Lowe, A., Locke, J., 2008. Enterprise resource planning and the post bureaucratic organization: “Formalization” as trust in the system versus “solidarity” as trust in individuals. Inform. Technol. People 21 (4), 375–400. Lu, Y., Ramamurthy, K., 2011. Understanding the link between information technology capability and organizational agility: an empirical examination. MIS Quart. 35 (4), 931–954. Menor, L.J., Roth, A.V., 2008. New service development competence and performance: an empirical investigation in retail banking. Prod. Oper. Manage. 17 (3), 267–284. Microsoft, 2013. TechNet: Microsoft Dynamics AX 2012: Application Users, http://technet.microsoft.com/en-us/library/hh242087.aspx (accessed 27.5.14). Michael, G., Astorian, J., Willford, M., 2012. Why big systems are here to stay. Accenture Outlook 14 (1), 1–8. Miller, D., Friesen, P.H., 1983. Strategy-making and environment: the third link. Strategic Manage. J. 4 (3), 221–235. Miller, J.W., Stromeyer, W.R., Schwieterman, M.A., 2013. Extensions of the Johnson–Neyman technique to linear models with curvilinear effects: derivations and analytical tools. Multivar. Behav. Res. 48 (2), 267–300. Mithas, S., Ramasubbu, N., Sambamurthy, V., 2011. How information management capability influences firm performance. MIS Quart. 35 (1), 237–256. Morton, N.A., Hu, Q., 2008. Implications of the fit between organizational structure and ERP: a structural contingency theory perspective. Int. J. Inform. Manage. 28 (5), 391–402. Nahm, A.Y., Vonderembse, M.A., Koufteros, X.A., 2003. The impact of organizational structure on time-based manufacturing and plant performance. J. Oper. Manage. 21 (3), 281–306. Nash, K.S., 2010. Reviving ERP. CIO 23 (7), 36–45. Olhager, J., Selldin, E., 2003. Enterprise resource planning survey of Swedish manufacturing firms. Eur. J. Oper. Res. 146 (2), 365–373. Oracle, 2013. Oracle E-Business Suite 12.2 Documentation Library, http://docs. oracle.com/cd/V39571 01/current/html/docset.html (accessed 27.5.14). Panorama Consulting, 2011. ERP Report. Panorama Consulting Group, Denver, Co. Patel, P.C., 2011. Role of manufacturing flexibility in managing duality of formalization and environmental uncertainty in emerging firms. J. Oper. Manage. 29 (1–2), 143–162. Pavlou, P.A., El Sawy, O.A., 2010. The “third hand”: IT-enabled competitive advantage in turbulence through improvisational capabilities. Inform. Syst. Res. 21 (3), 443–471. Peng, D.X., Schroeder, R.G., Shah, R., 2008. Linking routines to operations capabilities: a new perspective. J. Oper. Manage. 26 (6), 730–748. Platt, J.R., 1964. Strong inference. Science 146 (3642), 347–353. Podsakoff, P.M., MacKenzie, S.B., Lee, J.-Y., Podsakoff, N.P., 2003. Common method biases in behavioral research: a critical review of the literature and recommended remedies, Journal of. Appl. Psychol. 88 (5), 879–903. Podsakoff, P.M., Organ, D.W., 1986. Self-reports in organizational research: problems and prospects. J. Manage. 12 (4), 531–544. Popper, K.R., 1959. The Logic of Scientific Discovery, English ed. Basic Books, New York, NY. Preacher, K.J., Curran, P.J., Bauer, D.J., 2006. Computational tools for probing interaction effects in multiple linear regression, multilevel modeling, and latent curve analysis. J. Educ. Behav. Statist. 31 (4), 437–448. Prouty, K., Castellina, N., 2011. To ERP or Not to ERP: In Manufacturing, It Isn’t Even a Question. Aberdeen Group, Boston, MA. Rai, A., Hornyak, R., 2013. The impact of sourcing enterprise system use and work process interdependence on sourcing professionals’ job outcomes. J. Oper. Manage. 31 (6), 474–488. Ranganathan, C., Brown, C.V., 2006. ERP investments and the market value of firms: toward an understanding of influential ERP project variables. Inform. Syst. Res. 17 (2), 145–161. Rettig, C., 2007. The trouble with enterprise software. Sloan Manage. Rev. 49 (1), 21–27. Robey, D., Ross, J.W., Boudreau, M.-C., 2002. Learning to implement enterprise systems: an exploratory study of the dialectics of change. J. Manage. Inform. Syst. 19 (1), 17–46. Rosenzweig, E.D., 2009. A contingent view of e-collaboration and performance in manufacturing. J. Oper. Manage. 27 (6), 462–478. Rosenzweig, E.D., Roth, A.V., 2004. Towards a theory of competitive progression: evidence from high-tech manufacturing. Prod. Oper. Manage. 13 (4), 354–368. Rosenzweig, E.D., Roth, A.V., Dean, J.W., 2003. The influence of an integration strategy on competitive capabilities and business performance: an

164


exploratory study of consumer products manufacturers. J. Oper. Manage. 21 (4), 437–456. Rungtusanatham, M., Forza, C., Koka, B.R., Salvador, F., Nie, W., 2005. TQM across multiple countries: convergence hypothesis versus national specificity arguments. J. Oper. Manage. 23 (1), 43–63. Saeed, K.A., Abdinnour, S., Lengnick-Hall, M.L., Lengnick-Hall, C.A., 2010. Examining the impact of pre-implementation expectations on post-implementation use of enterprise systems: a longitudinal study. Decis. Sci. 41 (4), 659–688. Sambamurthy, V., Bharadwaj, A., Grover, V., 2003. Shaping agility through digital options: reconceptualizing the role of information technology in contemporary firms. MIS Quart. 27 (2), 237–263. SAP, 2013. SAP Help Portal: SAP ERP 6.0: Production Planning and Control, http://help.sap.com/saphelp erp60 sp/helpdata/en/ba/df293581dc1f79e 10000009b38f889/frameset.htm (accessed 27.5.14). Sarker, S., Sarker, S., Sahaym, A., Bjørn-Andersen, N., 2012. Exploring value cocreation in relationships between an ERP vendor and its partners: a revelatory case study. MIS Quart. 36 (1), 317–338. Sasidharan, S., Santhanam, R., Brass, D.J., Sambamurthy, V., 2012. The effects of social network structure on enterprise systems success: a longitudinal multilevel analysis. Inform. Syst. Res. 23 (3), 658–678. Scott, J.E., 2011. The paradoxical impact of enterprise-wide integration on flexibility. In: Kumar, S., Esteves, J., Bendoly, E. (Eds.), Handbook of Research in Enterprise Systems. Sage Publications, New Delhi, India, pp. 129–144. Scott, W.R., 2002. Organizations: Rational, Natural and Open Systems, 5th ed. Prentice Hall, Upper Saddle River, NJ. Schmenner, R.W., 2012. Getting and Staying Productive: Applying Swift, Even Flow to Practice. Cambridge University Press, Cambridge, UK. Schmenner, R.W., Vastag, G., 2006. Revisiting the theory of production competence: extensions and cross-validations. J. Oper. Manage. 24 (6), 893–909. Schroeder, R.G., Linderman, K., Liedtke, C., Choo, A.S., 2008. Six Sigma: definition and underlying theory. J. Oper. Manage. 26 (4), 536–554. Seddon, P.B., Calvert, C., Yang, S., 2010. A multi-project model of key factors affecting organizational benefits from enterprise systems. MIS Quart. 34 (2), 305–328. Setia, P., Patel, P.C., 2013. How information systems help create OM capabilities: consequents and antecedents of operational absorptive capacity. J. Oper. Manage. 31 (6), 409–431. Shah, R., Ward, P.T., 2007. Defining and developing measures of lean production. J. Oper. Manage. 25 (4), 785–805. ˜ G., 2011. Shaikh, N., Prabhu, V., Abril, D., Sánchez, D., Arias, J., Rodríguez, E., Riano, Kimberly-Clark Latin America builds an optimization-based system for machine scheduling. Interfaces 41 (5), 455–465. Shaw, J.D., Gupta, N., Delery, J.E., 2005. Alternative conceptualizations of the relationship between voluntary turnover and organizational performance. Acad. Manage. J. 48 (1), 50–68. Sia, S.K., Soh, C., 2007. An assessment of package-organisation misalignment: institutional and ontological structures. Eur. J. Inform. Syst. 16 (5), 568–583. Siemsen, E., Roth, A., Oliveira, P., 2010. Common method bias in regression models with linear, quadratic, and interaction effects. Organ. Res. Methods 13 (3), 456–476. Simon, H.A., 1962. The architecture of complexity. Proc. Am. Philos. Soc. 106 (6), 467–482. Sine, W.D., Mitsuhashi, H., Kirsch, D.A., 2006. Revisiting Burns and Stalker: formal structure and new venture performance in emerging economic sectors. Acad. Manage. J. 49 (1), 121–132. Singhal, V.R., 2008. Commentary to the editorial: empirical elephants—why multiple methods are essential to quality research in operations and supply chain management. J. Oper. Manage. 26 (3), 345. Snider, B., da Silveira, G.J.C., Balakrishnan, J., 2009. ERP implementation at SMEs: analysis of five Canadian cases. Int. J. Oper. Prod. Manage. 29 (1), 4–29.

Spear, S., Bowen, H.K., 1999. Decoding the DNA of the Toyota Production System. Harv. Bus. Rev. 77 (5), 96–106. Statistics Finland, 2010. Statistics of Science, Technology and Information Society: Innovation – Table 13, http://www.stat.fi/til/inn/2008/inn 2008 2010-06-10 tau 013 en.html (accessed 27.5.14). Stratman, J.K., 2007. Realizing benefits from enterprise resource planning: does strategic focus matter? Prod. Oper. Manage. 16 (2), 203–216. Strong, D.M., Volkoff, O., 2010. Understanding organization–enterprise system fit: a path to theorizing the information technology artifact. MIS Quart. 34 (4), 731–756. Swink, M., Nair, A., 2007. Capturing the competitive advantages of AMT: designmanufacturing integration as a complementary asset. J. Oper. Manage. 25 (3), 736–754. Swink, M., Narasimhan, R., Kim, S.W., 2005. Manufacturing practices and strategy integration: effects on cost efficiency, flexibility, and market-based performance. Decis. Sci. 36 (3), 427–457. Swink, M., Narasimhan, R., Wang, C., 2007. Managing beyond the factory walls: effects of four types of strategic integration on manufacturing plant performance. J. Oper. Manage. 25 (1), 148–164. Taylor, F.W., 1911. The Principles of Scientific Management. Harper & Brothers, New York, NY. Teece, D.J., 2007. Explicating dynamic capabilities: the nature and microfoundations of (sustainable) enterprise performance. Strategic Manage. J. 28 (13), 1319–1350. Teece, D.J., Pisano, G., Shuen, A., 1997. Dynamic capabilities and strategic management. Strategic Manage. J. 18 (7), 509–533. Tenhiälä, A., 2011. Contingency theory of capacity planning: the link between process types and planning methods. J. Oper. Manage. 29 (1–2), 65–77. Tenhiälä, A., Salvador, F., 2014. Looking inside glitch mitigation capability: the effect of intraorganizational communication channels. Decis. Sci. 45 (3), 437–466. Terjesen, S., Patel, P.C., Covin, J.G., 2011. Alliance diversity, environmental context and the value of manufacturing capabilities among new high technology ventures. J. Oper. Manage. 29 (1–2), 105–115. Tiwana, A., Konsynski, B., 2010. Complementarities between organizational IT architecture and governance structure. Inform. Syst. Res. 21 (2), 288–304. Upton, D.M., Staats, B.R., 2008. Radically simple IT. Harv. Bus. Rev. 86 (3), 118–124. Wailgum, T., 2008. The tie that binds: can’t live with them, can’t live without them. CIO 21 (9), 48–52. Walton, E.J., 2005. The persistence of bureaucracy: a meta-analysis of Weber’s model of bureaucratic control. Organ. Stud. 26 (4), 569–600. Wang, N., Liang, H., Zhong, W., Xue, Y., Xiao, J., 2012. Resource structuring or capability building? An empirical study of the business value of information technology. J. Manage. Inform. Syst. 29 (2), 325–367. Weber, M., 1946. The Theory of Social and Economic Organization. The Free Press, New York, NY. Xue, Y., Liang, H., Wu, L., 2011. Punishment, justice, and compliance in mandatory IT settings. Inform. Syst. Res. 22 (2), 400–414. Yeh, C.-H., Xu, Y., 2013. Managing critical success strategies for an enterprise resource planning project. Eur. J. Oper. Res. 230 (3), 604–614. Yuan, K.-H., Bentler, P.M., 2000. Three likelihood-based methods for mean and covariance structure analysis with nonnormal missing data. Sociol. Methodol. 30 (1), 165–200. Zhang, D., Linderman, K., Schroeder, R.G., 2012. The moderating role of contextual factors on quality management practices. J. Oper. Manage. 30 (1–2), 12–23. Zollo, M., Winter, S.G., 2002. Deliberate learning and the evolution of dynamic capabilities. Organ. Sci. 13 (3), 339–351.

Performance effects of using an ERP system for ...

Performance effects of using an ERP system for ...

Suggest Documents

ERP System Effects - GUPEA

Using a KM framework to evaluate an ERP system ...

A SYSTEM DEVELOPMENT METHODOLOGY FOR ERP SYSTEM ...

“Who needs an ERP system anyway?” - Usc

An applicable approach for performance auditing in ERP

The Effect of ERP System Implementation on Business Performance ...

PMS Project Management System An ERP For Construction Industry ...

Models for Selecting an ERP System with ... - Semantic Scholar

A simulator prototype for an ERP system - DTU ETD

Models for Selecting an ERP System with ... - Semantic Scholar

Performance Optimization of ERP-Based BCIs Using Dynamic Stopping

A High-Performance Implementation of an IoT System Using DPDK

Adoption of ERP system: An empirical study of factors ...

The Effect of ERP Implementation CSFs on Business Performance: An ...

An Expert System for Performance Management

An Initial Total-System Performance Assessment for

Implicit effects of emotional contexts: An ERP study - Springer Link

The implementation and deployment of an ERP system - Amazon Web ...

Analyzing the Implementation of an ERP System by Self-Assessment ...

Review of an ERP System Supporting Lean ... - Springer Link

Performance of Keystroke Biometrics Authentication System Using ...

Performance Evaluation of Distributed System Using SPN

Performance Improvement of Leaf Identification System Using ...

using erp system to improve internal supply chain coordination