A fuzzy expert system for business management

https://doi.org/10.1016/j.eswa.2010.04.086Get rights and content

Abstract

Nowadays firms are required to reach high levels of specialisation in order to increase their competitiveness in complex markets. Knowledge management plays a fundamental role in this process as the correct implementation of strategies is determined by the information transfer and dissemination within the organisation. In this paper, a fuzzy expert system focused on increasing accuracy and quality of the knowledge for decision making is designed. A model based on fuzzy rules to simulate the behavior of the firms, is presented under the assumption of determined input parameters previously detected and an algorithm is developed to achieve the minimal structure of the model. The result is a fuzzy expert system tool, called ESROM, which gives valuable information to help managers to improve the achievement of the strategic objectives of the company. A main contribution of this work it that the system is general and can be adapted to different scenarios.

Introduction

Nowadays, firm managers are becoming aware of the need for information analysis tools in order to support business decisions in the current complex and turbulent business environment (Handfield & Melnyk, 1998). Competition in changing environments due to fast progression of technical advances turns competition on information into the main competitive parameter in order to prevent and anticipate changes in customer needs, technology, industry trends and other competition parameters (Wacker, 1998). Therefore, the evolution of business computing networking and client/server architectures are impelling utilization of shared information in a decision support context.

In this context, the use of Decision Support Systems (DSS) is increasing and becoming generalized (Hasuike and Ishii, 2009, Inaad, 2009, Sharma et al., 2010, Vigier and Terceño, 2008). Even though, the development of new algorithms involves a fast progression in accuracy of DSS (Demoulin, 2007, Wen et al., 2008), the use of new DSS techniques has been scarcely applied in the field of Operations Management (OM) (Garbolino and Taroni, 2002, Lotfi and Pegels, 1996). In fact, even though management information systems literature has broadly dealt with tools to assist in managerial decisions, the wide utility these systems generate for specific Operations Management (OM) decisions is not still generalized (Stenforsa et al., 2007).

However, the use of surveys based on questionnaires in OM research is widely extended for academics and practitioners in order to define constructs, dimensions and variables to enhance understanding of OM issues (Wacker, 1998). Statistical multivariable techniques have been intensively applied in empirical studies with different levels of reliability and validity (O’Leary-Kelly & Vokurka, 1998).

Different studies have analyzed the relationship between operations strategy and performance through the use of statistical analysis, as we can see in Arias-Aranda, 2002, Arias-Aranda, 2003, Arias-Aranda et al., 2001. These studies analyze the relationship between operations strategy and performance through flexibility as a moderating variable within the service setting of engineering consulting firms in Spain. Artificial Intelligence (AI) techniques are also used in this field, like Bayesian classifiers (Abad-Grau & Arias-Aranda, 2006), expert systems (Miah et al., 2009, Shiue et al., 2008), case-base reasoning (Li and Ho, 2009, Lin et al., 2007) and so on.

Under these conditions, the aim of this research is to combine two different approaches: the use of surveys based on questionnaires in OM research with current techniques of AI. In this paper we develop a fuzzy modelling mechanism which is capable of implementing four objectives: (i) representing the knowledge obtained in terms of natural language, (ii) expressing the results obtained from the questionnaires analysis in a way that can be easily understood by non-experts users through fuzzy logic, (iii) generating a rule base automatically from numeric and linguistic data, (iv) acting as simulator of output results according to different input conditions controlled by the user. In order to achieve these objectives, this paper proposes a fuzzy expert systems, called ESROM, which will help to manager to make decisions about the company by means of simulating actual situations.

The rest of the paper is organized as follows: Section 2 provides some preliminaries on the fundamental theoretical aspects underlying this paper: Operations Management and Expert Systems. In Section 3, the case study is presented. In Section 4, an automatic learning algorithm applied to a determined relationship of OM variables will be suggested. In Section 5, an algorithm to obtain the minimal sets of rules to make more understandable and efficient the system will be presented. In Section 6, the expert system tool (ESROM) will be described and a simulation of the previously generated based system fuzzy rules will be done. Finally, in Section 7 presents the conclusions and future research.

Section snippets

Background

This section provides some basic background about the topics covered in the paper: Section 2.1 presents a quickly overviews about OM and Section 2.2 refreshes the basic ideas in Expert Systems.

The case study

In this section we describe the case study in detail. The expert gave us collected data about the operations strategy, level of flexibility and performance from a sample of 71 engineering consulting firms in Spain. Three firm types (Civil, Industrial and Environmental) were considered covering most activities of Engineering Consulting Firms. A questionnaire was the technique used to obtain data for the study. Initially, a copy of the questionnaire was sent to 10 firms representing every

Learning fuzzy knowledge obtained from questionnaires

In this application-oriented paper an automatic learning algorithm based on Operations Management survey is suggested, in order to help manager to make decisions about the company.

Getting minimal sets of rules

The complexity of the models is measured by the different number of combinations to be managed by the algorithm. For example in operations strategy, the questionnaire has nine blocks (hence nine variable) and five membership functions or linguistic labels associated to each variable (Very Low (VL), Low (L), Medium (M), High (H), Very High (VH)), thus generating 59 = 1953125 possible rules. The complexity is O(mn), being n the number of variables and m the number of linguistic labels.

In Section 4.2

System simulation

As we have said, our model can be used to understand in an easy way the knowledge acquired through the questionnaires and simulating the behavior of the firm and facilitating the task of making decisions about the company. We can see what kind of variables generate changes in the operations strategy, flexibility and performance. In this section we are going to see in detail: how we can use ESROM and finally some advantages of the system simulations.

Conclusions and future works

In this paper, a new tool (ESROM) based on fuzzy logic to improve the knowledge of the relationships between variables has been developed. This tool has been applied to a real Operations Management research and it can help managers to simulate strategic environments to obtain valuable information about levels of strategy, flexibility and performance required in the operations management area. ESROM allows to the users modify the generated rules when new knowledge is acquired in the firm. In

References (40)

  • Shah Jahan Miah et al.

    A methodology to allow rural extension professionals to build target-specific expert systems for Australian rural business operators

    Expert Systems with Applications

    (2009)
  • S.W. O’Leary-Kelly et al.

    The empirical assessment of construct validity

    Journal of Operations Management

    (1998)
  • R.V. Ramasesh et al.

    Measurement of manufacturing flexibility: A value based approach

    Journal of Operations Management

    (1991)
  • A. Sharma et al.

    Sustainability and business-to-business marketing: A framework and implications

    Industrial Marketing Management

    (2010)
  • W. Shiue et al.

    A frame knowledge system for managing financial decision knowledge

    Expert Systems with Applications

    (2008)
  • H.P. Vigier et al.

    A model for the prediction of diseases of firms by means of fuzzy relations

    Fuzzy Sets and Systems

    (2008)
  • J.G. Wacker

    A definition of theory: Research guidelines for different theory-building research methods in operations management

    Journal of Operations Management

    (1998)
  • W. Wen et al.

    A mobile knowledge management decision support system for automatically conducting an electronic business

    Knowledge-Based Systems

    (2008)
  • L.A. Zadeh

    Fuzzy sets

    Information and Control

    (1965)
  • M. Abad-Grau et al.

    Operations strategy and flexibility: Modeling with Bayesian classifiers

    Industrial Management and Data Systems

    (2006)
  • Cited by (33)

    • Vessel traffic scheduling method for the controlled waterways in the upper Yangtze River

      2019, Ocean Engineering
      Citation Excerpt :

      Knowledge representation is the process of organizing the knowledge, and the method of production rule is adopted. Based on certain rules, the inference engine derives a conclusion from known facts and can be successfully applied in different domains (Arias-Aranda et al., 2010; Grahovac and Devedzic, 2010). In order to enhance the traffic management and ensure navigation safety in the controlled waterways, we try to find an optimal sorted sequence of vessels using the inference engine.

    • A novel and fast MIMO fuzzy inference system based on a class of fuzzy clustering algorithms with interpretability and complexity analysis

      2017, Expert Systems with Applications
      Citation Excerpt :

      This system incorporates data and experts' knowledge into fuzzy rule base which significantly reduces required effort and time. A fuzzy expert system is proposed by combining artificial intelligence and survey based questionnaires to assist managers to improve achievements of their companies (Arias-Aranda, Castro, Navarro, Sanchez, & Zurita, 2010). The data obtained from questionnaires sent to some companies are shaped as fuzzy rules which represent experts' knowledge in terms of natural language.

    • How can Taiwan create a niche in Asia's cruise tourism industry?

      2016, Tourism Management
      Citation Excerpt :

      In 1975, some scholars proposed the fuzzy logic and the related application means (Mandami and Assilian, 1975; Zadeh, 1975). So far, the fuzzy theory has been widely applied in the academic and practical fields, so it is usually integrated with the expert system to resolve practical problems (Arias-Aranda, Castro, Navarro, Sanchez, & Zurita, 2010). The membership function is the basic concept of the fuzzy theory, which is the only way to quantize the fuzzy sets and the precise mathematical method to analyze and process the fuzzy data.

    • Fuzzy extension of a classical function: An alternative to the knowledge base for modelling imprecise relations

      2015, Applied Mathematical Modelling
      Citation Excerpt :

      In such cases, fuzzy logic together with fuzzy set theory are efficient mathematical tools for the rigorous modelling of these imprecise relations, sets of fuzzy rules of the type IF/THEN being frequently used for this purpose [1]. The description of imprecise relations by means of sets of fuzzy rules has been widely applied to different fields in the industrial sector, such as business management [2,3], industrial manufacturing [4–6] and process control [7–9], among others. It has also been successfully used in medicine as a useful tool for the diagnosis of diseases [10–13].

    View all citing articles on Scopus
    View full text