Analytic Network Process (ANP) has been widely used in solving many complicated decision making problems. An approach to acquire the comprehensive weights of interval-valued Analytic Network Process (IV-ANP) is proposed. Firstly, the independent local...
Analytic Network Process (ANP) has been widely used in solving many complicated decision making problems. An approach to acquire the comprehensive weights of interval-valued Analytic Network Process (IV-ANP) is proposed. Firstly, the independent local interval-valued weights are obtained by solving the goal programming for interval-valued pairwise comparison matrices. Based on the interaction and feedback relationships among criteria and/or indices, an unweighted interval-valued supermatrix is formed. To avoid loss of information, an extreme value method is developed for acquiring the global weights of IV-ANP. Secondly, the alternatives are ranked on the basis of interval-valued preference order. Finally, with the interaction and feedback relationships between criteria and/or indices being considered, an index system for assessing the risk of dynamic alliance is presented. With the uncertain and inaccurate information during the evaluation process being considered, the proposed IV-ANP method is applied for risk assessment of dynamic alliance, and sensitivity analyses for some key risk factors are performed to show their impacts on the risk levels of alliance enterprises.