A neural network evaluation model for ERP performance from SCM perspective to enhance enterprise competitive advantage

https://doi.org/10.1016/j.eswa.2007.08.102Get rights and content

Abstract

Due to increasing global competition, many enterprises are aware of the benefits of Enterprise Resource Planning (ERP). While the external environments and alliance partnerships facing an enterprise are becoming more complex, executives should consider appropriate partners to enhance efficiency and performance of supply chain management (SCM) as well as to gain potential competitive advantages. This study constructs a conceptual model to evaluate the performance and competitive advantages associated with ERP from a SCM perspective. The resulting model can be used to assist an enterprise in evaluating the potential partnerships. The survey data was gathered from a transnational textile firm in Taiwan. The training and learning models were based on the strategic thrust theory and used the Back-Propagation Network (BPN) as an evaluation tool.

Introduction

The internal information systems of a traditional organization are usually orientated on a functional basis. This set-up does not encourage efficient departmental communication within the firm. Traditional information systems do not satisfy the information requirements of global logistic trends. Recently, there has been an emphasis on integrating a company’s internal and external activities to improve a firm’s competitive edge. This approach, when applied to the development of integrated information systems, has become a major thrust. Davenport (1998) stated that an integrated information system is a smart tool that can be used by a firm to solve problems associated with widely distributed information sources.

Enterprise Resource Planning (ERP) systems can integrate a firm’s internal information from a financial perspective, allowing finance, accounting, purchasing and other departments to acquire information in a timely manner. ERP emphasizes integration of the flow of information relating to the major functions of the firm. The broader and more complex the organization is, the more it requires integrating this information flow. When applying supply chain management (SCM), orders can be forecasted efficiently and correctly, stock costs for supply chain partners can be reduced, and a manufacturing schedule can be set to optimize manufacturing and supply time. Additionally, strategic alliance was developed to facilitate collaboration between firms (Forrest & Martin, 1992). It plays an important role in establishing a firm’s competitive advantage (Bowersox, 1990). SCM emphasizes close collaboration between supply chain partners and the building of a strong alliance in their joint strategic business focus. Therefore, SCM and a firm’s competitive advantage are closely linked.

Integrating SCM to an ERP system can facilitate information flow in the supply chain so that partners of the chain can streamline their operations and share information sources to provide timely and accurate services to their customers. Traditional methods to evaluate ERP performance are limited to the internal departments of the company and do not include supply chain partners. However, under the global competition, many companies strengthen their core competencies via selecting their good business partners (Hong, Park, Jang, & Rho, 2005). Moreover, Choy, Lee, and Lo (2003) suggest that improving supply chain execution is important for achieving a firm’s competitive advantage. Shin, Collier, and Wilson (2000) emphasized that a firm’s performance can be evaluated by one or more key competitive priorities. Therefore, the five strategic forces of the strategic thrust theory can be independent or linked (Wiseman, 1985), and may relate to SCM performance.

This study uses a case to construct a conceptual model for the performance evaluation of an extended ERP system from an SCM perspective. The Back-Propagation Network (BPN) is used as a tool to access tacit knowledge held by the firm’s employees and the ERP consultants. This knowledge can be used to evaluate the extended ERP systems that conform to the SCM performances. The goals of this paper are as follows:

  • (1)

    To access the tacit knowledge inherent in the case firm’s employees and its ERP consultants/experts through the model learning process.

  • (2)

    To construct a BPN model to support a firm in evaluating its extended ERP performance from an SCM perspective and to test the competitive advantages gained by the ERP system.

  • (3)

    To produce results that will be useful to a firm when selecting partners.

Section snippets

Strategic thrust theory

Porter (1985) used a value chain to analyze the operations of firms in reaching global optimization by coordinating activities. Porter identified five key forces that enable a firm to establish a long-term competitive advantage. His “Five Forces Theory” comprises of the bargaining power of suppliers; the bargaining power of buyers; the potential threat of new entrants; the threat of substitute products or services; and rivalry among existing firms.

Wiseman (1984) proposed a Strategic Thrust

The constructing procedures of conceptual model

A conceptual model is constructed to evaluate competitive advantage based on an SCM perspective after the subject firm has implemented an extended ERP system. There are four steps to obtain the results of this study. First, in-depth interviews are conducted individually with three reputable consultants each having at least seven years consulting experience in ERP. This interview established the relationship between the criteria used for the firm’s SCM performance and the competitive advantages

The case firm

The case firm used to produce short staple and copied hair products. Due to the limited demand for these products the firm entered into the business of textile weaving and now produces spun cotton to be woven into cloth for making clothes. Recently, the case firm expanded into international operations and invested in a factory in Mexico making ready-to-wear clothes. It has integrated American market channels and factory sites in Mexico and Asian areas. The case firm has constructed a complete

Reliability and validity analysis

To measure the reliability of a questionnaire, it is common to use Cronbach’s Alpha to measure the consistency of research variables. When Cronbach’s α value is greater than 0.7, it is acceptable (Nunnally, 1978). Hair, Anderson, Tatham, and Black (1998) also supported this perspective and proposed that the research variables should be rejected if the Cronbach’s α value is less than 0.35. The Cronbach’s α value in this study is 0.8522, which implies that this questionnaire has a high

Limitations

This study collected the training and learning data from a case firm, focusing on its executives. We realize, however, that only a few executives participate in all the business operations and the decision-making strategies in the firm. Furthermore, if the firm’s partners do not do business electronically, then the extended ERP cannot promote integral competitive advantages. In this case, the values would be lower for ERP performance. This phenomenon also supports the use of this study in

References (38)

  • B.B. Wong et al.

    Neural network applications in business: a review and analysis of the literature (1988-95)

    Decision Support Systems

    (1997)
  • APICS (American Production & Inventory Control Society). (1998). APICS dictionary. Ninth...
  • D.J. Bowersox

    The strategic benefits of logistic alliances

    Harvard Business Review

    (1990)
  • G. Bylinsky

    The challengers move in on ERP

    Fortune

    (1999)
  • L.P. Cascante et al.

    The impact of expert decision support systems on the performance of new employees

    Information Resources Management Journal

    (2002)
  • A.L. Comrey

    A first course in factor analysis

    (1973)
  • T.H. Davenport

    Putting the enterprise into the enterprise system

    Harvard Business Review

    (1998)
  • P.P. Dornier et al.

    Global operations and logistics: text and cases

    (1998)
  • L.M. Ellram

    A managerial guideline for the development and implement of purchasing partnerships

    International Journal of Purchasing and Materials Management

    (1991)

    • Cited by (48)

    • ERP Neural Network Inventory Control

      2017, Procedia Computer Science

    • Linguistic performance evaluation for an ERP system with link failures

      2014, Information Sciences
      Citation Excerpt :

      They also utilized the Tobit regression to investigate the relationship between efficiency scores and the ERP continuous investment based on the concept of total cost ownership. The third way are from heuristics such as the work of Chand et al. [4] who provided a balanced-scorecard based framework for valuing the strategic performance of an ERP system, the work of Yang et al. [19] who presented a performance-evaluation model of ERP implementation utilizing fuzzy measures, the work of Chang et al. [5] who constructed a conceptual model to evaluate the performance and competitive advantages associated with ERP from a supply chain management perspective, and the work of Lin et al. [13] who proposed a statistical method based on Parasuraman and a revised performance evaluation matrix to set up a standard performance upper and lower control limits in terms of Taguchi method and Shewhart control charts. For quantitative method, Chen and Lin [6,7] firstly proposed novel analytic methods for evaluating the realtime performance of ERP systems.

    • A holonic approach to flexible flow shop scheduling under stochastic processing times

      2014, Computers and Operations Research
      Citation Excerpt :

      The first input is the mean of the CPTVs for all the machine holons in a cluster holon. Number of single hidden layers: Generally one hidden layer is capable of approximating any function with a finite number of discontinuities [37,8]. Therefore, a BPN only consists of one hidden layer.

    • Flexible flow shop scheduling with stochastic processing times: A decomposition-based approach

      2012, Computers and Industrial Engineering
      Citation Excerpt :

      All the four inputs are normalized in the range of [0, 1] for input into a BPN. Number of single hidden layers: Generally one hidden layer is capable of approximating any function with a finite number of discontinuities (Chang et al., 2008; Ripley, 1996). Therefore, a BPN only consists of one hidden layer.

    • Rural Planning Evaluation Based on Artificial Neural Network

      2022, Computational and Mathematical Methods in Medicine

    • Neural Network Model for the Multiple Factor Analysis of Economic Efficiency of an Enterprise

      2021, Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
    View all citing articles on Scopus
    View full text
    Copyright © 2007 Elsevier Ltd. All rights reserved.