Logistics Performance Evaluation Based On Linear Programming System Identification Method

Since rational and comprehensive logistics performance evaluation can help enterprises to improve the operational efficiency of the logistics system and enhance the core competitiveness, logistics performance has become an important part of enterprise management and control process. However, studies about logistics in China is still in the rise phase and it hasn’t been taken seriously so that logistics and the management of logistics enterprises still have many problems which means the existing evaluation system and method is not comprehensive enough. For this reason, the thesis studied on the performance evaluation of logistics enterprises form the following aspects:The thesis studied on literatures about the logistics performance, summarized the impact indicators of the performance evaluation of logistics system and logistics evaluation methods and summed up a logistics enterprises performance evaluation index system that universally applicable for comprehensive service type of logistic enterprises including five kinds of indices that are the transportation management, warehouse management, inventory management, information management, and enterprise development management procedure. Based on this system, this thesis put forward a logistics system performance evaluation method based on linear programming system identification method. First, the thesis did some primary theoretical research on this method, built a production function model for logistic companies which could found out the contribution of each input in producing each output and probed the connection between efficient DEA decision-making-units and linear programming system identification efficient ones. Then this thesis applied the method into logistics performance evaluation and put forward an evaluation method which started from objective data and established input-output models for the five categories of indicators and modeled the system as a linear programming system. After identifying this system by calculating the technical consumption matrix the target coefficient vector, the model can further calculated the output efficiency of each indicator so that the logistic system can be evaluated horizontally, vertically and comprehensively. In addition, the linear programming system identification method can also be well suited for dynamic multi-access and multi-output logistics system and can come up with the improvement method for each invalid unit so that the method has great advantage in comparison with the traditional logistics performance evaluation methods. Furthermore, after deeper analysis of the method, to deal with the problem that simply concentrating on objective data may not be able to meet the complex logistics system in the real-world, the thesis come up with a method combined the analytic hierarchy method with the linear programming system identification method which can take the subjective experience into consideration. This comprehensive method can combine the output efficiency with weight of each index from the subjective experience and make up the disadvantages of the two methods very well. At last, for both logistics enterprises performance evaluation methods and the production function model, the thesis presented numerical illustrations to illustrate the applicability and the superiority of them.