Group multi-attribute decision model to partner selection in the formation of virtual enterprise under incomplete information

https://doi.org/10.1016/j.eswa.2009.01.015Get rights and content

Abstract

Partner selection is very important in the formation of a virtual enterprise. However, in many practical situations, during the partner selection process, the decision makers are generally unsure of their preferences because the information about the candidates and their performances are incomplete and uncertain. Therefore, two MADM (multi-attribute decision model) methods for group decision making with interval values are proposed to solve partner selection problem under incomplete information in this paper. The first method is a technique for order preference by similarity to ideal solution (TOPSIS) for group decision making based on deviation degree. The second method is a TOPSIS for group decision making based on risk factor. Finally, an illustrative example is used to demonstrate the feasibility and practicability of these two extended TOPSIS methods for group decision making with interval values. Results show that these two extended TOPSIS methods for group decision making with interval values can effectively deal with the partner selection problem under incomplete information.

Introduction

The virtual enterprise (VE) is a kind of new organizational mode in the manufacturing world. A VE is a temporary alliance of autonomous, diverse, and possibly geographically dispersed organizations that share skills or core competencies to meet a particular mission and exploit fast changing market opportunity (Park and Favrel, 1999, Lau et al., 2000, Wu and Su, 2005). Hence, A VE is a kind of dynamic alliance of member companies, which join to take advantage of a market opportunity. When the market opportunity passes, the VE will be dissolved (Ip, Huang, Yung, & Wang, 2003).

The life cycle of a VE has four stages: creation, operation, evolution and dissolution. A major issue in the formation of the VE is to select suitable partners in the process of the dynamic alliance establishing, and it has attracted much research attention recently. For example, Talluri and Baker (1996) proposed a two-phase mathematical programming approach to solve the partner selection problem in the formation of a VE, where the factors of cost, time and distance were considered. Wang, Ip, and Yung (2001) developed a fuzzy decision embedded genetic algorithm to get the partner selection solution with due date constraint in a VE where the sub-projects form a precedence network. Ip et al. (2003) described and modeled a risk-based partner selection problem, and a rule-based genetic algorithm was developed to solve the partner selection problem. Zeng, Li, and Zhu (2006) proved that the partner selection problem with a due date constraint in a VE is a NP-complete problem, and a nonlinear integer programming model for this partner selection problem was established.

However, these traditional methods are problematic in that they use crisp values to express the decision makers’ judgment on a comparison of candidates. In reality, during the partner selection process, the decision makers are generally unsure of their preferences because the information about the candidates and their performances are incomplete and uncertain. Furthermore, some of the decision attributes are subjective and qualitative, so the decision makers cannot easily express their judgment on the candidates (Mikhailov, 2002, Wang and Chen, 2007). Nevertheless, the decision makers can utilize interval values to express the information about the candidates and their performances under incomplete and uncertain information. It means that some new methods are needed to be proposed to solve partner selection problem with interval values.

On the other hand, when selecting partners for a market opportunity in a VE, many factors will be taken into consideration, such as cost, quality, trust, credit, delivery time, reliability and so on Wu and Su (2005). And several decision makers will take part in the evaluation of candidates. Obviously, partner selection in the formation of a VE is a kind of group multi-attribute decision making (MADM) problem, which refers to making preference decisions over the available candidates that are characterized by multiple, usually conflicting, attributes for each decision maker. So group MADM methods can be used to solve partner selection problem of a VE.

According to the above mentioned analysis, two group MADM methods with interval values in solving partner selection problem under incomplete information are proposed in this paper. The first method is a technique for order preference by similarity to ideal solution (TOPSIS) for group decision making based on deviation degree. The second method is a TOPSIS group decision making based on risk factor.

This paper is organized as follows. Section 2 presents the group MADM problem with interval values. In Section 3 a TOPSIS for group decision making based on deviation degree and a TOPSIS for group decision making based on risk factor are proposed to solve the group MADM problem with interval values. Section 4 presents an illustrative example and Section 5 summarizes the work of this paper.

Section snippets

Presentation of the problem

Assuming an enterprise gets a bid for a large project consisting of several sub-projects. The enterprise is not able to complete the whole project only by using its own capacity. Therefore, the enterprise has to select partners for the sub-projects. The partner selection problem in the formation of a VE is described as follows:

Let X={Xi|i=1,,n} be a finite set of possible candidates and A={Aj|j=1,,m} be a finite set of attributes according to which the desirability of a candidate is to be

Extended TOPSIS for group decision making applied into partner selection

Hwang and Yoon, 1981, Yoon and Hwang, 1985 developed a TOPSIS for multiple attribute decision making. TOPSIS is based on the concept that the chosen candidate should have the shortest distance from the positive ideal solution (PIS) and the farthest distance from the negative ideal solution (NIS) for solving a multi-attribute decision-making problem. In traditional TOPSIS, most of the input variables are assumed to be crisp numerical data (Saremi, Mousavi, & Sanayei, 2009). However, in reality,

Numerical example

In order to demonstrate the application of two extended TOPSIS methods for group decision making to partner selection in the formation of a new virtual enterprise (VE), we will consider an example where the core enterprise of the virtual enterprise has to select a partner for a sub-project. The partner selection decision is made on the basis of five main attributes including Cost, Time, Trust, Risk and Quality. Cost, Time and Risk are cost type, while Trust and Quality are benefit type.

There

Conclusion

A major issue in the formation of the VE is to select suitable partners in the process of the virtual enterprise establishing, and it has attracted much research attention recently. In reality, in the partner selection process, the decision makers are generally unsure of there preferences because the information about the candidates and their performances are incomplete and uncertain. Therefore, in this paper we study the partner selection problem with interval values in the formation of a new

References (17)

There are more references available in the full text version of this article.

Cited by (96)

View all citing articles on Scopus
View full text