An Approach to Multi-attribute Group Decision Making and Its Application to Project Risk Assessment

Multiple attribute decision making is an important part of modern decision science. It has been extensively applied to various problems. There are several methods for this decision making problem. Based on TOPSIS method and Gray relational analysis method, this paper presents a combined approach to the multi­attribute group decision making. Finally, an illustrative example of evaluating projects risk assessment is given to show the feasibility and effectiveness of the developed method. Index Terms—multi­attribute group decision making, TOPSIS, GRA, project risk assessment I. INTRODUCTION Multiple attribute decision making (MADM) problems are to select an optimal alternative from a finite number of feasible alternatives based on the features of each attributes with respect to every alternative. It has been extensively applied to various areas. Many researchers have presented several methods for MADM such as AHP, TOPSIS, VIKOR, GRA, etc [1­9]. TOPSIS (technique for order preference by similarity to an ideal solution) method was presented by Hwang and Yoon [2]. The basic principle is that the chosen alternative should have the shortest distance from the ideal solution and the farthest distance from the negative ideal solution. Gray relational analysis (GRA) method was originally developed by Deng and has been successfully applied in solving a variety of MADM problems [4][10]. The gray relational degree between ideal alternative and every comparability alternative is calculated. If a comparability alternative has the highest gray relational degree between the ideal alternative and itself, that alternative will be the best choice. GRA method only considers the shape similarity of data sequence curve of alternative’s attribute to that of ideal solution’s [4]. However, TOPSIS method only considers the position approximation [11]. Combining TOPSIS and GRA method, we present a combined approach that can accurately reflect the relationship between alternative’s data and ideal solution’s. With the development of modern society, the socioeconomic environment has been becoming more complex, more multi­attribute decision making processes are determined by group decision makers [12­13]. Problems in engineering project risk is ubiquitous and no methods to avoid, but can be prevented and controlled as long as taking appropriate measures. Under certain conditions, we can farthest reduce the risk of loss. Engineering project risk evaluation is the important content of the engineering project risk management. For the comprehensive evaluation, we firstly identify risk factors and risk events or the specific parts, and then employ the related industry experts evaluate the risk factors or risk event. Finally, we use the multi­attribute group decision making approach to determine the overall risk of the project. The rest of this paper is structured as follows. In the next section, we present a brief introduction of TOPSIS method. In Section 3, Gray relational analysis (GRA) method is introduced. Combining TOPSIS and GRA method, we present a combined decision making approach in Section 4. For multi­attribute group decision making, an approach is given in Section 5. A numerical example given to illustrate the proposed method is shown in Section 6. II. TOPSIS METHOD Before we describe the procedure of TOPSIS method, let us assume a decision matrix, A , be defined as