Designing an expert system to support competitiveness through global sourcing

Designing an expert system to support competitiveness through global sourcing Authors’ pre-print version To cite this article: Ruggero Golini & Matteo Kalchschmidt (2015) Designing an expert system to support competitiveness through global sourcing, International Journal of Production Research, 53:13, 3836-3855, DOI: 10.1080/00207543.2014.974842 To link to this article: http://dx.doi.org/10.1080/00207543.2014.974842 Performing global sourcing – which is defined as the amount of purchases made by a plant outside its continent of origin – effectively is highly complicated because of several concurring variables (e.g., number of suppliers, supplied goods, and the location of the buyer and supplier, among other factors). However, the findings available in literature - which are often focused on few variables – can be difficult to be applied by managers to take decisions in this domain. Because of this limitation, we built an expert system (called GSES) to suggest what type of supply management (e.g., supplier base management, supplier selection, supplier integration) practices are most suitable given the level of global sourcing, strategy and other characteristics of the company. To build the theoretical grounding of the GSES we performed a deep literature review to identify the highest number of variables that can be related to a global sourcing strategy. Next, we built the GSES exploiting data from the International Manufacturing Strategy Survey as a knowledge base. This dataset is composed of 725 companies surveyed in 2009 in the assembly manufacturing industries (ISIC code range 2835) distributed across 19 countries. Finally, the system was validated by means of simulations and five real cases to verify that the expert system was actually usable by practitioners and provided robust insights. The paper also includes a detailed description of how the tool is built to allow future replications. Keywords: global sourcing strategy; expert system ; International Manufacturing Strategy Survey (IMSS); supply management

Acronyms used across the paper: • • • •

ARI: Average Reliability Index GSES: Global Sourcing Expert System HP: High performers IMSS: International Manufacturing Strategy Survey

• • •

1

JIT: Just in Time LP: Low performers SM: supply management

Introduction

Over time, globalization has led to an increase in the complexity of supply networks, which can affect firms’ performance if these networks are not properly managed. In this paper we focus on global sourcing defined as the practice of purchasing goods from countries located outside the continent where the plant is located. While opening new facilities abroad can require relevant financial efforts, global sourcing is also a viable strategy for smaller companies, and it is therefore more diffused (Quintens et al., 2005; Trent and Monczka, 2003). However, performing global sourcing effectively is anything but simple. When taking into account intercontinental transactions, global sourcing is not very diffused, and the effect on performance is not always positive (Cagliano et al., 2008; Caniato et al., 2013; Trent and Monczka, 2003). Indeed, several studies did not find any significant impact of global sourcing on companies’ performance (Kotabe and Omura, 1989; Steinle and Schiele, 2008). One of the challenges with global sourcing is that several concurring variables determine a successful global sourcing strategy. First, global sourcing needs to be managed appropriately. In other words, once the desired level of global sourcing is established, the key issue becomes the way in which the relationship with suppliers must be managed. The literature provides a broad set of leverages (i.e. tools and programs), such as sourcing strategies (e.g. single vs. multiple sourcing), supplier selection criteria, supplier development, use of electronic tools, and coordination and control over the supply chain (Flynn et al., 2010; Frohlich and Westbrook, 2001; Power, 2005). The problem is that different practices can impact the same performance, and bundles of practices can have complex effects. Therefore, it is difficult to isolate the effect of a single practice and to give precise suggestions to managers.

This picture is made even more complex if we consider that the effect of sourcing practices changes according to the context. For instance, it is important to take into account the characteristics of the company (e.g., company size), its environment (e.g., supply requirements variability, market dynamic) and its manufacturing system (e.g. product complexity, position of the decoupling point, and type of production) (Das and Handfield, 1997; Demeter and Golini, 2013; Kotabe and Murray, 2004; Sousa and Voss, 2008; Trent and Monczka, 2002; Zeng, 2003). Moreover, companies have different types of competitive priorities (e.g. cost, time, quality, flexibility, sustainability), which must also be taken into account to obtain from global sourcing what really supports the company strategy (Dyer and Singh, 1998; Ulaga, 2003). The complexity of the situation (i.e. many decisional variables, constraints and objectives) renders traditional approaches (e.g., focusing on few variables or practices at a time) (e.g., Holweg et al., 2011) not always suitable to support decision making in companies. To overcome such limitations, in this paper we built an expert system (called Global Sourcing Expert System – GSES) created with the aim to help theoretically any manufacturing firm to identify the most suitable supply management (SM) investments given a certain level of global sourcing and characteristics of the context. The application of expert systems to the domain of purchasing and SM is quite consolidated (e.g., Guneri et al., 2009; Ha and Krishnan, 2008; Ronchi et al., 2010; Wang et al., 2008), but no research in this stream has focused so far on global sourcing. However, in our view, only an expert system is able to consider simultaneously all of the complex relationships involved in a global sourcing strategy that are usually studied separately in the literature. In particular, the GSES is a knowledge-based expert system based on modelling and data mining (Liao, 2005). The system mines data collected in the 2009 edition of the International Manufacturing Strategy Survey (IMSS). This dataset is composed of 725 companies (in the ISIC code 28-35) distributed across 19 countries (for further information about the IMSS survey refer to the Appendixes). The limitation of the existing expert models is that they are usually mathematical or agentbased, and the literature generally lacks knowledge-based models in this area (e.g., Liao, 2005).

This gap is even more critical considering that several international surveys discuss SM topics (e.g. International Manufacturing Strategy Survey, International Purchasing Survey, High Performance Manufacturing), but none of them – to the best of our knowledge - has been ever used as a source of information for an expert system. For all these reasons, our contribution is threefold. First of all, we build a theoretical framework based on the literature that represents the main variables to be included in a global sourcing strategy and their relationships. Second, we show how to implement such framework in an expert system extracting useful knowledge from an international survey. Finally, to verify the reliability and applicability of the system, the GSES was tested with five different companies that provided useful insights for future developments. In the remainder of the paper, we will provide information on the relevant literature and the framework underpinning the methodology. Next, we will explain how the GSES was built and the results obtained from the case studies. Finally, we will draw the main conclusions.

2

Literature review

As an effect of globalization, the location of the suppliers and the identification of the most favourable opportunities for supply must take place on a global scale; otherwise, the company could be excluded from international markets and in many cases from the national competitive landscape as well (Gereffi and Lee, 2012). By exploiting the worldwide market, companies can realize benefits in terms of cost, quality, innovation, availability of materials and resources, access to new markets and new technologies (e.g., Bozarth et al., 1998; Holweg et al., 2011; Monczka and Trent, 1991; Nassimbeni, 2006; Zeng and Rossetti, 2003). However, companies should also pay attention to the issues that this type of strategy involves, e.g. the more difficult communications and relationship management with remote partners; more difficult management of economic, financial, information and material flows on wider spatial horizons; greater exposure to various risks and environmental

factors (e.g., Blackhurst et al., 2005; Chung et al., 2004; Deane et al., 2009; Golini and Kalchschmidt, 2011b; Manuj, 2013; Trent and Monczka, 2002; Zeng and Rossetti, 2003). In this paper, we refer to global sourcing as the purchases made outside the continent where the plant is based. A similar measure is used in Cagliano et al. (2008) and Caniato et al. (2013). According to the literature “global sourcing” implies high integration and coordination within the company and outside the company with its suppliers (Trent and Monczka, 2003; Zeng, 2003). This definition underlies a certain complexity in the management of this process. In fact, companies characterized by a high level of global sourcing are also more active in implementing internal and external (i.e. with suppliers) coordination mechanisms within a holistic global sourcing strategy (Golini and Kalchschmidt, 2011b).

2.1

The elements constituting a global sourcing strategy

According to the literature, a global sourcing strategy involves several elements. Together with the strategic indications and the main references, these elements are reported in Table 1. First, the number of suppliers is considered to be a relevant variable. Having many suppliers allows a company to keep suppliers in competition and to avoid under-capacity issues, but this strategy can reduce integration and suppliers’ responsiveness (e.g., Choi and Krause, 2006; Handfield et al., 2000). The criteria for the selection of suppliers are also very important as suppliers must be chosen according to the company’s competitive priorities (e.g., Choi and Krause, 2006; Lawson et al., 2009). Next, the way in which the suppliers are managed and integrated must be considered (e.g., Bartlett et al., 2007; Vereecke and Muylle, 2006). Furthermore, the level of visibility and the information sharing between the company and its suppliers is severely compromised by the implementation of global sourcing and must be managed accordingly (Chung et al., 2004; Dehning et al., 2007). In particular, information and communication technologies help, at least partially, to address the problem. Currently, concerning the purchasing-side, companies are

using several types of Internet-based SM tools. These are reported in Table 1 and later considered in our research. In addition, to implement such practices, additional programs are often needed. Before performing integration with suppliers, it is often necessary to review the supply strategy (e.g., Chan and Kumar, 2007; Ogden and Carter, 2008). Moreover, when a company creates closer ties with a few suppliers, it is also important to maintain some degree of control. Considerable emphasis has been placed on quality (Tan et al., 1999), especially in global SCs (Bayo-Moriones et al., 2011). More recently, monitoring corporate social responsibility of suppliers has also become an issue (e.g., Carter and Rogers, 2008; Seuring et al., 2008). Finally, global sourcing requires careful and onerous risk monitoring. It becomes crucial to analyse and carefully manage risk factors and provide proper contingencies and mitigation plans to address such risks (Juttner et al., 2003; Tang, 2006).

TABLE 1 ABOUT HERE

2.2

Internal and external factors related to a global sourcing strategy After the identification of the relevant elements that are part of a global sourcing strategy, it

is important to identify the factors that are related to such elements. A global sourcing strategy, in fact, must fit a set of factors that can be external or internal to the company (Bozarth and McDermott, 1998; Pun, 2005; Sarmiento et al., 2008). For example, Ward and Duray (2000) found that high-performing companies have a superior fit to the external and internal context. Table 2 summarizes the main internal and external factors identified in the literature review.

TABLE 2 ABOUT HERE

Starting from the external factors, product complexity was found to be a relevant variable affecting sourcing decisions (Kotabe and Murray, 2004; Sharon and Eppinger, 2001). Next, environmental uncertainty can appear in different ways, specifically according to the different causes that can introduce variability. In this work, attention is mainly given to the variability of demand and shortterm supply requirements (e.g., Lee, 2002). Concerning company variables, company size is one of the most relevant variables identified in the literature (Cagliano et al., 2008; Cavusgil, 1980; Lee and Whang, 2000; Quintens et al., 2005). Concerning production variables, the literature considers the position of the decoupling point (Naylor et al., 1999; Olhager and Östlund, 1990) and the type of production processes (one of a kind production, batch production, mass production) (e.g., Vickery, 1989). In addition, competitive priorities of a company should be considered. According to the literature, the term “competitive priorities” is mainly used to indicate “manufacturing tasks or key competitive capabilities, which broadly are expressed in terms of low cost, flexibility, quality, and delivery” (Kathuria, 2000).

3

Expert systems in supply management Table 1 points out the high number of elements constituting a global sourcing strategy while

Table 2 describes all the internal and external factors that should be considered in a global sourcing strategy. Interestingly, the literature provides contrasting evidence about several factors (e.g., product complexity, company size) witnessing the complexity of the topic and the difficulty to find solutions that are generally valid for every company. Given the difficulty in the definition of a global sourcing strategy, expert systems can provide a useful support. Expert systems are defined as “a computer program that represents and reasons with knowledge of some specialist subject with a view to solving problems or giving advice” (Jackson, 1999, p. 2). Expert systems and decision support systems have quite evolved over time (Eom, 1996), moving from individual to workgroup and virtual teams applications (Shim et al., 2002). Nevertheless, the individual tools - “usually small-scale systems that are developed for one manager, or a small number of independent

managers, to support a decision task”(Arnott and Pervan, 2008) - constitute the majority of the systems developed and they are considered those with higher impact. One key aspect in the scientific research about expert systems regards their rigorousness (Arnott and Pervan, 2008; Eom, 1999) as suggested by the many papers focused on validation and verification of expert systems (Hamilton et al., 1991; O'Keefe and O'Leary, 1993). The application of expert systems to the domain of SM is quite consolidated (Jayaraman and Srivastava, 1996). Expert systems in SM are typically used to support decisions in sourcing (e.g., Ronchi et al., 2010) and distribution (e.g., Adenso-Díaz et al., 1998; Faulin et al., 2005). In particular, in the domain of purchasing and SM, several authors (Guneri et al., 2009; Ha and Krishnan, 2008; Vokurka et al., 1996) developed expert systems for supplier evaluation and selection, Wang et at. (2008) for on-demand e-supply chain integration; Ronchi et al.(2010) for evaluating e-Procurement systems. However, to the best of our knowledge, no expert system has been created so far to support global sourcing decisions.

4

Objectives and Methodology Given the mentioned gaps in the literature, our objective is to design an expert system to

identify the best supply management practices that, in line with the level of global sourcing, the context and the competitive priorities, allow the achievement of superior performance. To fulfil this aim, we used the literature-based framework (reported in Figure 1) where the level of global sourcing, the internal and external factors constitute the input, the SM practices are the output, and performance is a control variable to identify the best and worst performers. In particular, SM practices are those described in Table 1 while internal and external factors are defined as in Table 2.

FIGURE 1 ABOUT HERE

The same logic was used to build the GSES. This expert system includes the aforementioned variables and theoretical relationships to allow different types of companies to identify the optimal way to support their global sourcing strategy through SM practices. It is important to emphasize that the model was developed considering the plant as the unit of analysis. We are aware that in multiplant networks, a broader set of strategies to deploy a global sourcing strategy exists than the one considered in our paper. Nevertheless, the GSES can also be employed by companies operating in multi-plant networks or by headquarters to orient their strategy. The GSES was also validated through simulations and five case studies. A detailed description of the GSES and the steps followed to build it is provided in the next paragraphs.

4.1

Design of the Global Sourcing Expert System (GSES) According to the literature, the process of conception and implementation of an expert

system follows four main stages: problem selection (1), in which the domain and boundaries of the problem are set; knowledge representation and acquisition (2), in which the information is collected and organized; knowledge engineering (3), which is the phase of real expert system development that includes: choice of the technology to employ, variable selection, realization of the inference engine and the presentation layer (the user interface) of the expert system; and finally, knowledge testing and evaluation (4) to test the compliance of the expert system in relation to its specifications and to evaluate and validate it in the actual use context. In our case, the problem domain and the relationships among variables were drawn from the literature, while the information came from the International Manufacturing Strategy Survey (IMSS) 5. This dataset is composed of 725 companies (in the ISIC code 28-35) distributed across 19 countries. This survey was already used to explore SM issues (e.g., Caniato et al., 2013; Golini and Kalchschmidt, 2011b; Wiengarten et al., 2014) but never in the form of an expert system. For more information about the IMSS database refer to the Appendixes.

It is important to remind that to build the GSES we did not use the results of the survey, but the GSES uses raw data to produce results that vary dynamically according to the inputs. This knowledge engineering phase required to create an Excel-based application that works along 4 steps, in which only the first and the last steps were visible to the user (Figure 2). In order to allow replications of our study, we describe how the GSES was build following each step.

FIGURE 2 ABOUT HERE

1.1.1 Step 1: Input user form First, the users inserts the inputs to create a profile of his/her company. As specified in Figure 1, the inputs regard the adoption of global sourcing and the internal and external factors. A detailed list of the inputs and measures is reported in Table 3. Measures (e.g., percentages, 1-5 scores) were selected in order to be aligned with the analogous questions in the database. From the system point of view, it is just a form, no calculations are made here. TABLE 3 ABOUT HERE

1.1.2 Step 2: Definition of the peer list The IMSS database used includes a large number of companies that have very different characteristics in terms of internal and external factors (Table 2). Therefore it is necessary for the user company to compare itself with a subset of companies with similar characteristics (i.e., peer list). Because of that, after the inputs are filled in, a short-list of peers is selected from the IMSS database (Step 2 in Figure 2). To be included in the peer list, the level of dissimilarity of a company in the database should be below a certain threshold. The level of dissimilarity is evaluated according to two indexes. The first one is the dissimilarity in terms of extent of global sourcing adoption (“global sourcing dissimilarity index”); the second one is the dissimilarity in terms of internal and external factors (“factors dissimilarity index”). There is an AND logic that connects the two

indexes. In other words, only companies that have similar global sourcing adoption and similar internal and external factors are included in the peer list. To evaluate the global sourcing dissimilarity index (idgs,i), we used information from the global sourcing section of the inputs (Table 3) where the respondent is asked to fill in the percentage of purchases from inside the nation (pin), outside the nation but inside the continent (pic), and outside the continent (poc). In the input form, the sum of the three components is required to be equal to 100%. We refer to a user company u; therefore, for each company i in the database, we calculate the global sourcing dissimilarity index for of company i as follows:

Every indicator (pin, pic, poc) is expressed in percentage terms (from 0% to 100%) and their sum must be 100%. Because of that, this index can range from 0% to 200%. When idgs,i is equal to 0% there is a perfect similarity, i.e. the user company and the company i in the dataset have the same sourcing profile (e.g., they both buy 100% outside the continent). When idgs,i is equal to 200% there is a complete dissimilarity, i.e., the user company and the company i in the dataset have an opposite sourcing profile (e.g., one buys 100% inside the country and the other one 100% outside). For the reasons explained later, we decided to include in the peers group only companies with an idgs,i value lower than 100%. Next, the factors dissimilarity index (IDi) is calculated as the average of several sub-indexes representing production type (idtp,i), position of the decoupling point (idpd,i), size measured with the turnover (idturn,i), size measured with the employees (idempl,i), market dynamic (iddyn,i),and variability of supply (idvar,i), requirement product complexity (idcompl,i), as follows:

The calculation of each sub index is reported in the annexes for brevity sake. Only the cases with a value of IDi lower than 120% (out of 200%) can enter in the peer group.

In conclusion, for each company in the database two dissimilarity indexes are calculated based on global sourcing and internal/external factors. Only those companies that are considered to be similar to the user company according to both dissimilarity indexes (global sourcing and general) are then included in the sample of peers. The threshold to include or not include a company in the peers list was calculated by balancing the need to find similar companies with the need to keep a minimum number of peers in the list to allow statistical analyses. Using the aforementioned thresholds, through a random generation of the inputs we established that, out of 725 companies in the database, the minimum peer list sample is made of 43 peers, the maximum of 350 and the average sample is made of 211 peer companies. We verified that such numbers allow to keep a sufficient statistical power for the further analyses.

1.1.3 Step 3: Definition of the high-performer and low-performer groups

Thereafter (Step 3), peers are divided into high and low performers (HP and LP). To evaluate performance, we defined a performance index as an average of several performance indicators (e.g., cost, delivery, quality, flexibility, and sustainability) (e.g.,Wiengarten et al., 2011) weighted by the importance declared by the user in the competitive priorities section of the inputs (Table 3). In particular, the performance indicators were the results of a factor analysis (the items are reported in brackets): cost (manufacturing cost; procurement cost), quality (product quality and reliability; manufacturing conformance), innovativeness (product innovativeness), delivery (delivery speed; delivery reliability), flexibility (volume flexibility; mix flexibility), service (customer service and support), sustainability (social reputation; environmental performance) and customization (product customization). For more details about the factor analysis, see Table A1 in the annexes.

As previously mentioned, the performance index (pi) is calculated for each company (i) as the weighted average of its performance indicator k (price, quality, etc.) times the score of the corresponding indicator in the competitive priorities section (sk,u) of the inputs provided by the user.

Accordingly, it is possible to identify companies that scored higher on the performance indicators that actually matter for the user. In particular, HP companies are identified among those having a performance index strictly higher than 3 out of 5. The scale went from 1 (much worse than the competitors) to 5 (much better than the competitors). Therefore, using a threshold of 3, we consider as HP only those companies that have a performance index strictly higher than that of the competitors. At this stage, the GSES assigns to each company in the peer list a binary value to identify HPs and LPs.

1.1.4 Step 4: Presentation of the output Finally (Step 4), the GSES presents a graphical and numerical representation of the outputs (i.e., a list of SM practices) specifically distinguishing what HP and LP companies do and providing a statistical significance indicator of the differences. In practical terms, for each SM practice, the GSES calculates two values of mean and standard deviation separately for the peers that are HP and LP. Next a Student t test with nHP + nLP – 2 degrees of freedom is performed using the following formula: 𝑡=

µ!" − µ!" 𝑛!" − 1 ∗ 𝜎!" ! + 𝑛!" − 1 ∗ 𝜎!" ! 𝑛!" + 𝑛!" ∗ 𝑛 ∗𝑛 𝑛!" + 𝑛!" − 2 !" !"

,

where: µHP e µLP are the average values for HP and LP, σ2HP and σ2LP are the variances, nHP e nLP are the number of HP and LP in the peer list. From the calculated t-value it is possible to determine the significance in percentage terms (p-value).

As previously mentioned, this calculation is performed for each SM practice. For a complete list of the outputs and an example of the graphical result, see Table 6 and Figure A1 in the annexes. In order to facilitate the understanding, the GSES indicates with a semaphore light indicator the degree of significance (green 10%).

4.2

GSES validation Following the recommendations of the literature (Hamilton et al., 1991; Knauf et al., 2002;

Vokurka et al., 1996), we performed two subsequent types of validation: the first one using simulation and the second one using real cases. For the first type of validation, after setting up the model, we performed several simulations to verify that the GSES provided results that were aligned with the both literature and our expectations. For these simulations, we used several cases drawn from the same IMSS database. This phase was very helpful in determining that the results were reasonable before using the GSES with the companies and to fine-tune the system. Next, the GSES was tested with real companies - also known as case testing (O'Keefe and O'Leary, 1993). According to the literature, since expert systems emulate human knowledge, it is important to verify it with real people (Knauf et al., 2002). Following the suggestion of the literature (O'Leary et al., 1990; Preece, 1990), given the early stage of development of this expert system (it is the first one about global sourcing using an international survey as a knowledge base), we performed a qualitative testing i.e., a subjective evaluation of the performance (Borenstein, 1998). To this aim, we interviewed five companies (from small to large), and we applied the GSES to their case. The companies were selected to represent a wide variety of characteristics (e.g. different size, level of global sourcing, competitive priorities), but they all belonged to the industrial sectors of the companies in the IMSS database. More details on these cases are reported in Table 4.

The operation or purchasing managers were interviewed. The interviews usually began by asking for a brief presentation of the firm (particularly in terms of the purchasing organization and approach to global sourcing). Next, the inputs of the GSES were completed, and the outputs were interpreted together. Finally, managers were asked for feedback on critical points, deficiencies and suggestions to make the GSES more usable without the support of the researchers. TABLE 4 ABOUT HERE In order to present the results of the qualitative evaluation, a parameter called the “reliability index” was calculated for each case. This approach is similar to the one used by Borenstein et al.(1998) that employed a questionnaire filled by the assessors to provide an assessment of the system reliability. In our case, the reliability index is a score given by the assessors to every variable in the analysis. The value was 1 if the indication provided by the GSES for that variable was congruous with the firm’s expectations, 0 if it was not and 0.5 if the company neither agreed nor disagreed. The reliability indexes were averaged for each set of SM practices (so to obtain a “section reliability index”) and then the score for each section was averaged into a “case reliability index”. Finally, averaging the values of all case studies, we obtained an Average Reliability Index (ARI), which provides a synthetic description of the GSES’s global trustworthiness. The interviews also helped to elucidate some variables and relationships that were not included in the GSES, which might be integrated in future developments.

5 5.1

Results and discussion Simulation

Before using the GSES with companies, we performed several simulations to verify that the GSES provided results aligned with both the literature and with our expectations. For these simulations, we used several cases drawn from the same IMSS database. This phase was very helpful for determining that the results were reasonable before using the GSES with the companies and for

fine-tuning the system (for instance, adjusting the similarity index to avoid the exclusion of too many companies from the peers list). To exemplify this process, we present here the results obtained for an illustrative case in the IMSS database. The case is a business unit of a large Italian company in the aeronautic sector with a fairly high level of global sourcing (40%). The inputs provided to the GSES about this case are reported in Table 5. TABLE 5 ABOUT HERE For this case, the GSES identified 201 peer companies, of which 76 were High Performers (HP) and 125 were Low Performers (LP) based on the importance of the declared competitive priorities. In this case, as all of the competitive priorities had the same weight, HP companies were those that had a higher average performance. The outputs provided by the GSES are reported in Table 6 and described below. TABLE 6 ABOUT HERE Regarding the number of suppliers, LP companies had more suppliers in absolute terms (239 vs. 214) and more key suppliers (31 vs. 26) compared to HP companies. This result is in agreement with Chen et al. (2004), and all of the studies in the literature show that global sourcing is favoured by having a limited number of selected suppliers with whom tight communication and collaboration is fostered despite the geographical distance, language and cultural differences (Choi and Krause, 2006; Humphreys et al., 2004; Li et al., 2006; Li et al., 2007; Li et al., 2000; Schoenherr, 2010). The criteria for selecting suppliers were very similar and no major differences were found between HP and LP companies. Delivery and quality were the most important criteria, followed by price and supplier potential. This result is in line with the literature that states that price is an important factor for global sourcing, but quality and delivery problems can nullify these savings (Handfield, 1994). These issues are so relevant that HP and LP companies behaved very similarly.

We found a similar outcome for information sharing and supplier integration, where HP and LP companies had, on average, a high level of adoption. In particular, order tracking and tracing was one of the most widespread practices, which makes perfect sense when freights move globally. Interestingly, however, only HP companies had a significantly higher adoption of Just-In-Time (JIT). JIT-global sourcing is still quite controversial in the literature (Das and Handfield, 1997; Samli et al., 1998; Vickery, 1989), and our results support that global JIT can significantly contribute to performance. HP companies are likely able to implement JIT thanks to assembly-toorder production, which allows these companies to have a lean supply chain upstream and an agile supply chain downstream (Naylor et al., 1999). Concerning the electronic means for interfacing with key/strategic suppliers, HP companies were particularly active in using these means for research and pre-selection of suppliers, for the analysis of data and for the management of orders and contracts (in agreement with Chung et al., 2004 and Dehning et al., 2007). These tools support communication when suppliers are located far away from one another. Moreover, the literature reports that global sourcing companies are particularly active in scouting for new suppliers and opportunities (Barbieri et al., 2011). Platforms for suppliers’ qualifications can be another tool to receive information from the supply market and to check that ethical standards are met. As discussed in the next paragraph, HP companies were also more active in controlling social issues along the supply chain. Contracts can be another critical element of global relationships, and the use of contract management tools can simplify and standardize contract execution. Finally, regarding SM improvement programs, HP companies differed significantly from LP companies regarding a higher implementation of supplier development and adoption of supply risk management (Chopra and Sodhi, 2004; Deane et al., 2009; Juttner et al., 2003; Prater et al., 2001; Sheffi, 2001). Finally, HP companies invested significantly more in social responsibility along the supply chain thanks to the support provided by other SM practices (Darnall et al., 2008; Gualandris

et al., 2014). In conclusion, a company such as that in the analysis should prioritize the implementation of the following SM practices and programs: (1) Keep a small supplier base; (2) Keep delivery and quality as the most important selection criteria, followed by price and supplier potential; (3) Implement order tracking and tracing, agree on delivery frequency and, when possible, adopt JIT; (4) Massively rely on electronic sourcing tools; (5) Invest in advanced SM programs, such as risk management, CSR monitoring and supplier development.

5.2

Testing cases

In the second phase, as described in the methodology, we collected data from five companies and applied the GSES on their cases in collaboration with the managers . The general perception is that the GSES provides reasonable answers to the companies. In other words, companies are usually aligned with what HP companies do or they recognize directions for improvement in their practices. To quantify this perception – as mentioned in the methodology section of this paper – we calculated a reliability index that measured how much the GSES provided outputs that are consistent with the expectations of each company. Table 7 shows the outcomes of the evaluation in terms of the reliability for each company and the total average.

TABLE 7 ABOUT HERE

We find that the ARI is 0.75, which indicates that, on average, 75% of the items considered by the GSES are aligned with the company’s expectations (i.e. 26 of 35 variables were consistently evaluated). This outcome represents a very good result, considering the heterogeneity of the data

about the companies gathered in the IMSS database in terms of industry, production characteristics, size, strategy and context. Interestingly, there are some areas in which the GSES was more effective and others that instead need some refinements. The main deviations found can be summarized according to the six GSES outputs. Number of suppliers. In this section, the GSES fit well for all of the companies interviewed; no specific points of disagreement were found among the GSES and the companies. Suppliers’ selection criteria. This is the lower-performing part of the GSES. This problem can be related to the fact that the GSES considers direct material purchases in aggregate without a distinction among different purchasing categories that, in turn, could be related to different supplier selection criteria. Supplier integration (information sharing). This part performed quite well except for Case 2 that considers information sharing risky and not as much relevant as suggested by the GSES. However, we consider this attitude not in line with the best practices identified by the literature so we do not consider this outcome an issue for the reliability of the GSES. Supplier integration (joint investments). This part performed quite well, except for one output: “Just-in-time.” Again, the GSES suggested a higher adoption, but companies were very careful about accepting this suggestion. Often, companies argued that although JIT represented a useful technique, it was difficult to implement to a level that produced significant results. Electronic tools. In this case, the GSES fit fairly well, and the only exception was represented by the tool “Order management and tracking”, which was not strongly considered by the interviewed companies. Since this tool is quite widespread in the IMSS sample, we deem that the bias can stand in the sample rather than in the GSES. In particular, Case 1 and 3 showed a lower than average adoption of electronic tools probably due to the smaller size of these companies. Supply management improvement programs. The GSES provided good results in this area. Surprisingly, there was quite high agreement even on sophisticated programs such as risk management, monitoring quality and social responsibility of the suppliers.

6

Conclusions Decision support systems, have undergone a process of critical review about their practical

usefulness and theoretical contribution (Eom, 1999). As a consequence, Arnott et al. (2008), among other things, suggest to undertake more case studies, develop rigorous research and to pick problems that are relevant for the management. Following this indications, in this paper we addressed a current managerial problem that is the definition of a global sourcing strategy (Roland Berger, 2011) and we rigorously developed and tested an expert system (GSES) to support such decisions. To confirm this, almost all of the interviewees agreed on the innovativeness and the sturdiness of the basic framework and the consequent usefulness of the instrument developed. Moreover, the fact that few inputs were solicited and that filling in the inputs did not require more than 15 minutes was highly appreciated by the managers. The GSES showed to be a good support to take decisions like: given my current level of global sourcing, what type of investments or interventions in the SM should my company undertake to become more competitive? Or - in perspective - my company wants to increase the level of global sourcing: what type of investments or interventions in the SM should my company undertake? Despite the good results achieved by the project both from the empirical and theoretical point of view, the system is not exempt from criticism and possible improvements. One key idea that emerged was the possibility to let managers use the GSES without the researchers’ support. From this standpoint, it will be important to reduce the use of some terminology that sometimes is perceived to be too academic or too technical, thereby leading to an unclear understanding of the inputs and outputs. For instance, it was not very clear whether the percentage of “domestic,” “continental” and “international” purchases should be expressed in terms of purchase volume or value. Moreover, it required further explanations to let user know how to behave in a situation of (1) supply by a local supplier that had manufacturing plants in foreign countries or (2) supplies from

a foreign supplier that sold its products through local dealers. Another question raised by the users was whether product complexity referred to all products of the company or one specific product. Lastly, the GSES took into account direct materials in general, but specifying this information at the category level could increase the interpretability of the results. The GSES has also some structural limitations. First of all, the IMSS database used in the GSES collects companies belonging to the assembly manufacturing industry (ISIC code range 2835), therefore its applicability to other industries (e.g. textile, food, etc.) is questionable and should be carefully evaluated. Finally, companies with very unusual input characteristics could result in having a very small number of peers, thus lowering the statistical power of the outputs. Concerning the paper’s contributions, similar results would be generated using a clustering technique to identify groups on the basis of inputs and to check the differences in the output variables. In the literature, some authors have followed this approach (Cousins et al., 2006; Terpend et al., 2011). Nevertheless, the number of input variables was so high that applying this approach would result in a very high number of clusters and in losing the possibility to tailor the instrument to the user’s specific case. For instance, Terpend et al. (2011) created clusters of different buyersupplier relationships and correlated them with different performance indicators, showing that the more strategic the relationship, the higher the performance. However, the GSES was able to weight the performance on the strategic objectives of the company, thus suggesting only those areas of investment that are more profitable. Building an expert system like the GSES allows us, in fact, to consider simultaneously all of the complex relationships that are usually studied separately in the literature. Our contribution, finally, lies not only in the results obtained, but also in showing how to turn a survey created with scientific purposes into an expert system that is able to involve companies and provide them with useful insights. Believing in this approach, we will try to address the aforementioned limitations in future releases of the expert system. However, we deem our approach, even at this stage, to be a step forward not only for the practitioners, but also for the scientific community.

7

References

Adenso-Díaz, B., González, M., Garcia, E., 1998. A hierarchical approach to managing dairy routing. Interfaces 28 (2), 21-31. Arnott, D., Pervan, G., 2008. Eight key issues for the decision support systems discipline. Decision Support Systems 44 (3), 657-672. Barbieri, P., Gavazza, L., Prodi, G., 2011. Supply China Management. Il Mulino, Bologna. Bartlett, P.A., Julien, D.M., Baines, T.S., 2007. Improving supply chain performance through improved visibility. International Journal of Logistics Management 18 (2), 294-313. Bayo-Moriones, A., Bello-Pintado, A., Merino-Díaz-de-Cerio, J., 2011. Quality assurance practices in the global supply chain: the effect of supplier localisation. International Journal of Production Research 49 (1), 255-268. Blackhurst, J., Craighead, C.W., Elkins, D., Handfield, R.B., 2005. An empirically derived agenda of critical research issues for managing supply-chain disruptions. International Journal of Production Research 43 (19), 4067-4081. Borenstein, D., 1998. Towards a practical method to validate decision support systems. Decision Support Systems 23 (3), 227-239. Bozarth, C., Handfield, R., Das, A., 1998. Stages of global sourcing strategy evolution: an exploratory study. Journal of Operations Management 16 (2-3), 241-255. Bozarth, C., McDermott, C., 1998. Configurations in manufacturing strategy: a review and directions for future research. Journal of Operations Management 16 (4), 427-439. Cagliano, R., Caniato, F., Golini, R., Kalchschmidt, M., Spina, G., 2008. Supply chain configurations in a global environment: A longitudinal perspective. Operations Management Research 1 (2), 86-94. Cagliano, R., Caniato, F., Spina, G., 2003. E-business strategy: how companies are shaping their supply chain through the internet. International Journal of Operations & Production Management 23 (10), 11421162. Caniato, F., Cagliano, R., Kalchschmidt, M., Golini, R., Spina, G., 2009. Evolutionary patterns in e-business strategy. International Journal of Operations & Production Management 29 (9), 921-945. Caniato, F., Golini, R., Kalchschmidt, M., 2013. The effect of global supply chain configuration on the relationship between supply chain improvement programs and performance. International Journal of Production Economics 143 (2), 285-293. Carter, C.R., Rogers, D.S., 2008. A framework of sustainable supply chain management: moving toward new theory. International journal of physical distribution & logistics management 38 (5), 360-387. Cavusgil, S.T., 1980. On the internationalization process of firms. European Research 8 (6), 273-281. Chan, F.T.S., Kumar, N., 2007. Global supplier development considering risk factors using fuzzy extended AHP-based approach. Omega 35 (4), 417-431. Chen, I.J., Paulraj, A., Lado, A.A., 2004. Strategic purchasing, supply management, and firm performance. Journal of Operations Management 22 (5), 505-523. Choi, T.Y., Krause, D.R., 2006. The supply base and its complexity: Implications for transaction costs, risks, responsiveness, and innovation. Journal of Operations Management 24 (5), 637-652. Chopra, S., Sodhi, M.M.S., 2004. Managing risks to avoid supply-chain breakdown. MIT Sloan management review (Fall 2004) 46 (1).

Chung, W.W.C., Yam, A.Y.K., Chan, M.F.S., 2004. Networked enterprise: A new business model for global sourcing. International Journal of Production Economics 87 (3), 267-280. Cousins, P.D., Lawson, B., Squire, B., 2006. An empirical taxonomy of purchasing functions. International Journal of Operations & Production Management 26 (7), 775-794. Darnall, N., Jolley, G.J., Handfield, R., 2008. Environmental management systems and green supply chain management: complements for sustainability? Business Strategy and the Environment 17 (1), 30-45. Das, A., Handfield, R.B., 1997. Just-in-time and logistics in global sourcing: an empirical study. Management 27 (3/4), 244-259. Deane, J.K., Craighead, C.W., Ragsdale, C.T., 2009. Mitigating environmental and density risk in global sourcing. International Journal of Physical Distribution & Logistics Management 39 (10), 861-883. Dehning, B., Richardson, V.J., Zmud, R.W., 2007. The financial performance effects of IT-based supply chain management systems in manufacturing firms. Journal of Operations Management 25 (4), 806-824. Demeter, K., Golini, R., 2013. Inventory configurations and drivers: An international study of assembling industries. International Journal of Production Economics (in press). Dyer, J.H., Singh, H., 1998. The relational view: Cooperative strategy and sources of interorganizational competitive advantage. Academy of management review 23 (4), 660-679. Eom, S.B., 1996. Mapping the intellectual structure of research in decision support systems through author cocitation analysis (1971–1993). Decision Support Systems 16 (4), 315-338. Eom, S.B., 1999. Decision support systems research: current state and trends. Industrial Management & Data Systems 99 (5), 213-221. Faulin, J., Sarobe, P., Simal, J., 2005. The DSS LOGDIS optimizes delivery routes for FRILAC’s frozen products. Interfaces 35 (3), 202-214. Fine, C.H., Hax, A.C., 1985. Manufacturing strategy: a methodology and an illustration. Interfaces 15 (6), 28-46. Fisher, M.L., 1997. What is the right supply chain for your product? Harvard Business Review 75, 105-117. Flynn, B.B., Huo, B., Zhao, X., 2010. The impact of supply chain integration on performance: A contingency and configuration approach. Journal of Operations Management 28 (1), 58-71. Frohlich, M.T., Westbrook, R., 2001. Arcs of integration: An international study of supply chain strategies. Journal of Operations Management 19 (2), 185-200. Gereffi, G., Lee, J., 2012. Why the world suddenly cares about global supply chains. Journal of Supply Chain Management 48 (3), 24-32. Golini, R., Kalchschmidt, M., 2011a. Managing inventories in global sourcing contexts: a contingency perspective, 20th IPSERA Conference, Maastricht (The Netherlands). Golini, R., Kalchschmidt, M., 2011b. Moderating the impact of global sourcing on inventories through supply chain management. International Journal of Production Economics 133 (1), 86-94. Gualandris, J., Golini, R., Kalchschmidt, M., 2014. Do supply management and global sourcing matter for firm sustainability performance? An international study. Supply Chain Management: An International Journal 19 (3). Gunasekaran, A., Ngai, E.W.T., 2005. Build-to-order supply chain management: a literature review and framework for development. Journal of Operations Management 23 (5), 423-451. Guneri, A.F., Yucel, A., Ayyildiz, G., 2009. An integrated fuzzy-lp approach for a supplier selection problem in supply chain management. Expert Systems with Applications 36 (5), 9223-9228. Ha, S.H., Krishnan, R., 2008. A hybrid approach to supplier selection for the maintenance of a competitive supply chain. Expert Systems with Applications 34 (2), 1303-1311.

Hamilton, D., Kelley, K., Culbert, C., 1991. State-of-the-practice in knowledge-based system verification and validation. Expert Systems with Applications 3 (4), 403-410. Handfield, R., Krause, D., Scannell, T., Monczka, R.M., 2000. Avoid the pitfalls in Supplier Development. Sloan Management Review 41 (2), 37-49. Handfield, R.B., 1994. US global sourcing: patterns of development. International Journal of Operations & Production Management 14, 40-40. Hayes, R.H., Wheelwright, S.C., 1984. Restoring our competitive edge: Competing through manufacturing. John Wiley & Sons Inc. Holweg, M., Reichhart, A., Hong, E., 2011. On risk and cost in global sourcing. International Journal of Production Economics 131 (1), 333-341. Humphreys, P.K., Li, W.L., Chan, L.Y., 2004. The impact of supplier development on buyer-supplier performance. Omega 32 (2), 131-143. Jackson, P., 1999. Introduction to Expert Systems, Third Edition ed. Pearson Education Limited, Edinburgh Gate, UK. Jayaraman, V., Srivastava, R., 1996. Expert systems in production and operations management: current applications and future prospects. International Journal of Operations & Production Management 16 (12), 27-44. Juttner, U., Peck, H., Christopher, M., 2003. Supply chain risk management: outlining an agenda for future research. International Journal of Logistics: Research and Applications 6 (4), 197-210. Kathuria, R., 2000. Competitive priorities and managerial performance: a taxonomy of small manufacturers. Journal of Operations Management 18 (6), 627-641. Knauf, R., Gonzalez, A.J., Abel, T., 2002. A framework for validation of rule-based systems. Systems, Man, and Cybernetics, Part B: Cybernetics, IEEE Transactions on 32 (3), 281-295. Kotabe, M., Murray, J.Y., 2004. Global sourcing strategy and sustainable competitive advantage. Industrial Marketing Management 33 (1), 7-14. Kotabe, M., Omura, G.S., 1989. Sourcing Strategies of European and Japanese Multinationals: A Comparison. Journal of International Business Studies 20 (1). Krause, D.R., Handfield, R.B., Scannell, T.V., 1998. An empirical investigation of supplier development: reactive and strategic processes. Journal of Operations Management 17 (1), 39-58. Lawson, B., Cousins, P.D., Handfield, R.B., Petersen, K.J., 2009. Strategic purchasing, supply management practices and buyer performance improvement: an empirical study of UK manufacturing organisations. International Journal of Production Research 47 (10), 2649-2667. Lee, H.L., 2002. Aligning supply chain strategies with product uncertainties. California management review 44 (3), 105-119. Lee, H.L., Whang, S., 2000. Information sharing in a supply chain. International Journal of Manufacturing Technology and Management 1 (1), 79-93. Li, S., Ragu-Nathan, B., Ragu-Nathan, T.S., Subba Rao, S., 2006. The impact of supply chain management practices on competitive advantage and organizational performance. Omega 34 (2), 107-124. Li, W., Humphreys, P.K., Yeung, A.C.L., Edwin Cheng, T.C., 2007. The impact of specific supplier development efforts on buyer competitive advantage: an empirical model. International Journal of Production Economics 106 (1), 230-247. Li, Z.G., Murray, L.W., Scott, D., 2000. Global sourcing, multiple country-of-origin facets, and consumer reactions. Journal of Business Research 47 (2), 121-133. Liao, S.H., 2005. Expert system methodologies and applications a decade review from 1995 to 2004. Expert systems with applications 28 (1), 93-103.

Lin, C., Chow, W.S., Madu, C.N., Kuei, C.H., Pei Yu, P., 2005. A structural equation model of supply chain quality management and organizational performance. International Journal of Production Economics 96 (3), 355-365. Luzzini, D., Longoni, A., Moretto, A., Caniato, F., Brun, A., 2013. Organizing IT purchases: Evidence from a global study. Journal of Purchasing and Supply Management. Manuj, I., 2013. Risk Management in Global Sourcing. Transportation Journal 52 (1), 80-107. Monczka, R.M., Trent, R.J., 1991. Global sourcing: a development approach. International Journal of Purchasing and Materials Management 27 (2), 2-8. Nassimbeni, G., 2006. International sourcing: Empirical evidence from a sample of Italian firms. International Journal of Production Economics 103 (2), 694-706. Naylor, B.J., Naim, M.M., Berry, D., 1999. Leagility: integrating the lean and agile manufacturing paradigms in the total supply chain. International Journal of Production Economics 62 (1), 107-118. O'Keefe, R.M., O'Leary, D.E., 1993. Expert system verification and validation: a survey and tutorial. Artificial Intelligence Review 7 (1), 3-42. O'Leary, T.J., Goul, M., Moffitt, K.E., Radwan, A.E., 1990. Validating expert systems. IEEE Expert 5 (3), 51-58. Ogden, J.A., Carter, P.L., 2008. The supply base reduction process: an empirical investigation. International Journal of Logistics Management, The 19 (1), 5-28. Olhager, J., Östlund, B., 1990. An integrated push-pull manufacturing strategy. European Journal of Operational Research 45 (2-3), 135-142. Perona, M., Miragliotta, G., 2004. Complexity management and supply chain performance assessment. A field study and a conceptual framework. International Journal of Production Economics 90 (1), 103-115. Power, D., 2005. Supply chain management integration and implementation: a literature review. Supply Chain Management: An International Journal 10 (4), 252-263. Prater, E., Biehl, M., Smith, M.A., 2001. International supply chain agility-Tradeoffs between flexibility and uncertainty. International Journal of Operations & Production Management 21 (5/6), 823-839. Preece, A.D., 1990. Towards a methodology for evaluating expert systems. Expert Systems 7 (4), 215-223. Pun, K.F., 2005. An empirical investigation of strategy determinants and choices in manufacturing enterprises. Journal of Manufacturing Technology Management 16 (3), 282-301. Quintens, L., Matthyssens, P., Faes, W., 2005. Purchasing internationalisation on both sides of the Atlantic. Journal of Purchasing and Supply Management 11 (2-3), 57-71. Roland Berger, 2011. Purchasing Excellence 2011. Ronchi, S., Brun, A., Golini, R., Fan, X., 2010. What is the value of an IT e-procurement system? Journal of Purchasing and Supply Management 16 (2), 131-140. Ruamsook, K., Russell, D.M., Thomchick, E.A., 2009. Sourcing from low-cost countries: Identifying sourcing issues and prioritizing impacts on logistics performance. International Journal of Logistics Management 20 (1), 79-96. Samli, A.C., Browning, J.M., Busbia, C., 1998. The Status of Global Sourcing as a Critical Tool of Strategic Planning:: Opportunistic Versus Strategic Dichotomy. Journal of Business Research 43 (3), 177-187. Sarmiento, R., Knowles, G., Byrne, M., 2008. Strategic consensus on manufacturing competitive priorities: A new methodology and proposals for research. Journal of Manufacturing Technology Management 19 (7), 830-843. Schoenherr, T., 2010. Outsourcing decisions in global supply chains: an exploratory multi-country survey. International Journal of Production Research 48 (2), 343-378.

Seuring, S., Sarkis, J., Müller, M., Rao, P., 2008. Sustainability and supply chain management - an introduction to the special issue. Journal of Cleaner Production 16 (15), 1545-1551. Sharon, N., Eppinger, S.D., 2001. Sourcing by Design: Product Complexity and the Supply Chain. Management Science 47 (1), 189-204. Sheffi, Y., 2001. Supply chain management under the threat of international terrorism. International Journal of Logistics Management, The 12 (2), 1-11. Shim, J.P., Warkentin, M., Courtney, J.F., Power, D.J., Sharda, R., Carlsson, C., 2002. Past, present, and future of decision support technology. Decision support systems 33 (2), 111-126. Sousa, R., Voss, C.A., 2008. Contingency research in operations management practices. Journal of Operations Management 26 (6), 697-713. Steinle, C., Schiele, H., 2008. Limits to global sourcing? Strategic consequences of dependency on international suppliers: Cluster theory, resource-based view and case studies. Journal of Purchasing and Supply Management 14 (1), 3-14. Tan, K.C., Kannan, V.R., Handfield, R.B., Ghosh, S., 1999. Supply chain management: an empirical study of its impact on performance. International Journal of Operations & Production Management 19 (10), 10341052. Tang, C.S., 2006. Perspectives in supply chain risk management. International Journal of Production Economics 103 (2), 451-488. Terpend, R., Krause, D.R., Dooley, K.J., 2011. Managing buyer-supplier relationships: empirical patterns of strategy formulation in industrial purchasing. Journal of Supply Chain Management 47 (1), 73-94. Trent, R.J., Monczka, R.M., 2002. Pursuing competitive advantage through integrated global sourcing. The Academy of Management Executive (1993-2005), 66-80. Trent, R.J., Monczka, R.M., 2003. Understanding integrated global sourcing. International Journal of Physical Distribution & Logistics Management 33 (7), 607-629. Ulaga, W., 2003. Capturing value creation in business relationships: a customer perspective. Industrial Marketing Management 32 (8), 677-693. Vereecke, A., Muylle, S., 2006. Performance improvement through supply chain collaboration in Europe. International Journal of Operations & Production Management 26 (11), 1176-1198. Vickery, S.K., 1989. International sourcing: implications for just-in-time manufacturing. Production and Inventory Management Journal 30 (3), 66-71. Vokurka, R.J., Choobineh, J., Vadi, L., 1996. A prototype expert system for the evaluation and selection of potential suppliers. International Journal of Operations & Production Management 16 (12), 106-127. Wagner, S.M., Bode, C., 2006. An empirical investigation into supply chain vulnerability. Journal of Purchasing and Supply Management 12 (6), 301-312. Wang, M., Liu, J., Wang, H., Cheung, W.K., Xie, X., 2008. On-demand e-supply chain integration: A multiagent constraint-based approach. Expert Systems with Applications 34 (4), 2683-2692. Ward, P.T., Duray, R., 2000. Manufacturing strategy in context: environment, competitive strategy and manufacturing strategy. Journal of Operations Management 18 (2), 123-138. Westhead, P., Wright, M., Ucbasaran, D., 2001. The internationalization of new and small firms:: A resource-based view. Journal of Business Venturing 16 (4), 333-358. Wiengarten, F., Fynes, B., Pagell, M., de Búrca, S., 2011. Exploring the impact of national culture on investments in manufacturing practices and performance: an empirical multi-country study. International Journal of Operations & Production Management 31 (5), 554-578. Wiengarten, F., Pagell, M., Ahmed, M.U., Gimenez, C., 2014. Do a country's logistical capabilities moderate the external integration performance relationship? Journal of Operations Management 32 (1), 51-63.

Zeng, A.Z., 2003. Global sourcing: process and design for efficient management. Supply Chain Management: An International Journal 8 (4), 367–379. Zeng, A.Z., Rossetti, C., 2003. Developing a framework for evaluating the logistics costs in global sourcing processes. International Journal of Physical Distribution and Logistics Management 33 (9).

Figure 1 – The literature-based framework used to develop the expert system. In dark grey are the inputs and in light grey are the outputs of the methodology. Performance acts as a control variable.

Figure 2 – The logical flow of the GSES

Table 1 – SM practices and programs as elements of a global sourcing strategy

Element

Strategic indications

Number of suppliers

• • •

Supplier selection criteria

•

Keep many suppliers to increase competition and to avoid under-capacity issues Keep less suppliers to increase responsiveness Keep local back-up suppliers to face global sourcing risks Align supplier selection criteria (e.g., physical proximity, price, quality, flexibility, lead time, willingness to collaborate) to the strategy

References Choi & Krause, 2006; Handfield et al., 2000

Choi & Krause, 2006; Lawson et al., 2009; Lin et al., 2005; Wagner & Bode, 2006

Supplier integration

Determine the right amount of collaboration and integration Bartlett, Julien & Baines, with suppliers in the production-logistics processes in terms 2007; Cagliano, Caniato & of: Spina, 2003; Frohlich & Westbrook, 2001; Ruamsook, • Information sharing: refers to the exchange of Russell & Thomchick, 2009; information about production plans, inventories, market Vereecke & Muylle, 2006 demand • Joint investments made with suppliers to coordinate physical activities (e.g. Just-in-Time, Vendor Managed Inventory) in order to achieve faster flows of products with less inventory levels Use of • Use Internet-based tools (e.g., RFx1 management Chung, Yam & Chan, 2004; electronic software, order management and tracking tools, Dehning, Richardson & tools reporting and decision support systems) to foster Zmud, 2007; Luzzini et al., communication and increase visibility especially with 2013; Caniato et al., 2009) suppliers that are far away Supply • Review the supply strategy (e.g. reduce the supply base, Chan & Kumar, 2007; management implement vendor ratings, and, in case the suppliers are Krause, Handfield & improvement not very skillful, investment in supplier development is Scannell, 1998; Ogden & programs Carter, 2008 needed) •

• 1

Supplier quality and sustainability monitoring (e.g. workplace safety, working conditions, harmful emissions and energy efficiency) Monitoring risks in the global supply chain and define mitigation plans

Carter & Rogers, 2008; Seuring et al., 2008 Juttner, Peck & Christopher, 2003; Tang, 2006

RFx include: Request for Information (RFI), Request for Proposal (RFP), Request for Quote (RFQ), and Request for Bid (RFB) sent from a customer to a supplier

Table 2 – Internal and external factors affecting a global sourcing strategy

Contextual factors and their connection with global sourcing

References

Ext. Product complexity

Chung, Yam & Chan, 2004; Golini & Kalchschmidt, The less complex the product, the easier it is to scout for and communicate with 2011; Kotabe & Murray, suppliers abroad. However, complex products may require to look for suppliers 2004; Perona & Miragliotta, abroad if they are not available locally 2004; Sharon & Eppinger, 2001; Westhead, Wright & Ucbasaran, 2001 Variability of demand and supply requirements

Fisher, 1997; Lee, 2002

When variability is significant, more integration with suppliers typically leads to better performance Int.

Company size

Cagliano et al., 2008; Cavusgil, 1980; Lee & To source globally and to implement SC management, both financial and human Whang, 2000; Quintens, resources are needed. Larger companies generally have more access to these Matthyssens & Faes, 2005 resources. However, some studies indicate that smaller companies can also adopt global sourcing significantly, even if they do not possess the necessary resources to effectively perform such sourcing Position of the decoupling point

Gunasekaran & Ngai, 2005

Companies that operate in make-to-stock contexts can be more efficient in managing their material inventory and consignment plans. Companies that operate in make-toorder contexts must be more reactive so that they either have to become more integrated with suppliers or have to maintain higher inventory levels Type of production processes (one of a kind production, batch production, mass Das & Handfield, 1997; production) Handfield, 1994; Samli, Browning & Busbia, 1998; Global sourcing seems to work badly with small lot sizes, but other authors support Vickery, 1989 the opposite idea, where global Just-In-Time (and thus lot sizes tending towards the unitary lot) can significantly contribute to businesses performance. Competitive priorities (quality, cost, flexibility, delivery, innovation) Implementing a global sourcing strategy involves advantages and difficulties that can have a significant impact on the possibility of supporting competitive priorities; on the other hand, implementing one business strategy rather than another affects the way in which global sourcing is deployed into SM practices

Fine & Hax, 1985; Hayes & Wheelwright, 1984

Table 3 - GSES Inputs

Section

Global sourcing

Question Percentage of sourcing (raw materials, parts/components, subassemblies/systems) from: • This country • Within your continent • Outside your continent Note: The sum of these three values must be 100%

Product complexity

Product complexity (1- Very few parts/materials, one-line bill of material; 5-Many parts/materials, complex bill of material)

Variability of demand and supply • • requirements Company Size

Competitive priorities

Type of production processes

Position of the decoupling point

Market dynamics (1-Declining rapidly; 5-Growing rapidly) Weekly supply requirements variability (1-Low; 5-High)

• Number of employees • Turnover 1-5 (from not important to very important) importance of the following attributes to win orders from major customers: • Price • Quality • Innovativeness • Delivery • Flexibility • Service • Sustainability • Customization Percentage of the production volumes: • One of a kind production • Batch production • Mass production Note: The sum of these three values must be 100% Percentage of customer orders that are: • Designed/engineered to order • Manufactured to order • Assembled to order • Produced to stock Note: The sum of these three values must be 100%

Table 4 – The case studies characteristics

Case 1

Main product Inside Italy Inside Europe Outside Europe Cost Quality Importance Delivery Flexibility of competitive Sustainability priorities Service Customization Innovativeness Size (n. of employees) Global sourcing

Type of production1

Context 1

Market Demand variability Product complex.

Case 2

Case 4 Case 5 Molds for cars and Textile machinery Hydraulic Boards and boats, machinery for industrial cranes and control electronic and Electronic material for rotational automation systems lighting equipment molding 80% 60% 80% 70% 10% 15% 40% 15% 27% 90% 5% 0% 5% 3% 0% Low Low Low Low Medium High High High High High High Medium Medium Medium Medium Low High Medium High Medium Low High Low Low Low High Medium Medium Low Medium High High High High Low High Low High High Medium 12 500 23 300 1600 Mainly MTS in Mainly ETO one of Mainly ETO by ETO in one of a MTO by batches continuous a kind production batches kind production production Mature Mature/Declining Mature/Growing Mature Mature Low High Average Average Average/Low High High Medium High High

ETO: Engineer to Order; MTO: Make to Order; MTS: Make to stock

Case 3

Table 5 – Illustrative case information (inputs)

Global sourcing 20% inside the country 40% in Europe 40% outside Europe

Competitive priorities Balanced on all the priorities

Size 1900 employees 270 M€ Turnover

Type of production Mainly Assembly to Order by batches

Context Mature market Low demand variability Very complex product

Table 6 - GSES Outputs for the illustrative case (statistics are based on 76 HP and 125 LP). Statistically significant differences (i.e. 10%

Order tracking/tracing

3.70 3.48 >10%

Dedicated capacity Vendor managed inventory or consignment stock Plan, forecast and replenish collaboratively Just-in-time replenishment (e.g. kanban) Physical integration Electronic Scouting/ pre-qualify tools Auctions (1-5 Scale) RFx (request for quotation, proposal, information) Data analysis (audit and reporting) Order management and tracking Contract and document management Supply Rethinking and restructuring supply strategy and the organization and management management of supplier portfolio through e.g. tiered networks, bundled outsourcing, and supply improvement base reduction programs Implementing supplier development and vendor rating programs (1-5 Scale) Increasing the level of coordination of planning decisions and flow of goods with suppliers including dedicated investments (e.g. information systems, dedicated capacity/tools/ equipment, dedicated workforce) Implementing supply chain risk management practices including early warning system, effective contingency programs for possible supply chain disruptions Monitoring corporate social responsibility of partners along the supply chain (e.g. labor conditions) Increasing the control of product quality along the supply chain (raw materials and components certification, supplier audit, product integrity in distribution, etc.)

LP 214 31 3.58 4.05 4.12 3.09 3.06 2.54 2.89 3.12

Sig. >10% >10% 10% >10% >10% >10% >10% >10% >10%

3.07 2.88 >10%

3.07 2.85 3.27 3.15 2.26 3.19 1.83 3.66 3.33 3.78 3.51

2.92 2.62 2.99 2.52 1.72 2.54 1.50 3.24 2.88 3.23 2.99

>10% >10% >10% 10% >10%