The Third-party Logistics Performance Evaluation Based On The AHP-PCA Model

As an advanced third-party logistics management technology and organization, it is widely regarded as the third profit source after the lower material consumption and the higher productivity. So how objectively, fairly and comprehensively to evaluate the third-party logistics enterprise performance relates to the direction, intensity and the quality of investment decisions which the business investors rebalance. Then according to the valid assessment, establish effective corporate incentive and restraint mechanisms which is conducive to the three logistics companies to find problems, to strengthen management, to improve operations, and to enhance core competitiveness. Therefore, the performance evaluation of the third party logistics problem has important theoretical and practical significance.The paper analyzes the definition of third party logistics and the characteristics based on the third-party logistics enterprise performance evaluation of the relevant content. Discuss the problem to exist for a long time which is about the third party logistics performance evaluation of a problem. Based on previous research results, propose an evaluation index system of third party logistics framework, which includes the financial performance indicators, customer management performance indicators, operational process performance indicators and learning and growth indicators. To target these four aspects of the index, performance evaluation of third party logistics.For the problem of the third party logistics performance evaluation, use the combination of qualitative analysis (AHP) and quantitative analysis (PCA) method. Using AHP to determine the weight coefficients of all indicators, and analyze indicators of principal component, and obtain the principal components; Weight the standardized index, and take into account the weights. Establish a mathematical model (AHP-PCA) to optimize third-party logistics performance evaluation. Finally, use an example to prove the model feasible and effective.