A parallel network DEA model based on a global perspective
Production systems with parallel network structure are a representative class of complex production systems.Since some important parallel network DEA models have not well dealt with the relationship between the decision making unit(DMU)and its internal subunits,this leads to the problems that the DEA projection points may overflow the production possibility set and the objective function may not accurately reflect the efficiency of DMUs.In order to solve these problems,this paper first analyzes the problems of the original parallel network DEA model through some examples.Then,a parallel network DEA model based on a global perspective is proposed,and the relevant properties of the model are discussed.Finally,the model of this paper is applied to the empirical analysis of forest production efficiency in Taiwan,China.The results show that the model of this paper can not only measure the efficiency of DMUs from a global perspective,give information on the adjustment of inefficient DMUs and their subunits,but also solve the problem of efficiency analysis with differences in the numbers of subunits.