Ehsan Vaezi
Abstract
Classic Data Envelopment Analysis methods ignore the internal interactions and consider the systems as a ‘black box’. Most of the network analysis models are nonlinear and it is feasible that these point may cause a considerable amount of modification to occur in the efficiency results. Amidst ...
Read More
Classic Data Envelopment Analysis methods ignore the internal interactions and consider the systems as a ‘black box’. Most of the network analysis models are nonlinear and it is feasible that these point may cause a considerable amount of modification to occur in the efficiency results. Amidst which, models, such as, the Wang Model, takes the intermediate variables into account, but in order to prevent intricacies in resolving models, it has an inconclusive approach, between two outlooks, of the black box and the network. Hence, in this paper, we consider a three-stage network with additional, desirable and undesirable inputs and outputs and the three abovementioned approaches are analyzed by contemplating on the optimistic and pessimistic views. The goals of this paper are to put together the results of the three mentioned approaches in order to attain the final conclusions. We generalize and use of Wang’s approach for the three levels, with additional inputs and outputs, as well as a heuristic solution to solve the network’s view. Finally, this paper considers a genuine world example, in the form of a dynamic network, for model application and analyzes it from three perspectives.
Ehsan Vaezi; Seyyed Esmaeil Najafi; Seyed mohamad Haji Molana; Farhad Hosseinzadeh Lotfi; Mahnaz Ahadzadeh Namin
Abstract
Data Envelopment Analysis (DEA) is one of the methods most widely used for measuring the relative efficiency of DMUs in the world today. The efficiency evaluation of the network structure opens the “black box” and considers the internal structure of systems. In this paper, a three-stage ...
Read More
Data Envelopment Analysis (DEA) is one of the methods most widely used for measuring the relative efficiency of DMUs in the world today. The efficiency evaluation of the network structure opens the “black box” and considers the internal structure of systems. In this paper, a three-stage network model is considered with additional inputs and undesirable outputs and obtains the efficiency of the network, as interval efficiency in presence of the imprecise datum. The proposed model of this paper simulates a factory in the factual world with a production area, three warehouses and two delivery points. This factory is taken into consideration as a dynamic network and a multiplicative DEA approach is utilized to measure efficiency. Given the non-linearity of the models, a heuristic method is used to linearize the models. Ultimately, this paper concentrates on the interval efficiency to rank the units. The results of this ranking demonstrated that the time periods namely, (24) and (1) were the best and the poorest periods, respectively, in context to the interval efficiency within 24 phases of time.