The System And Application Of Multi-attribute Comprehensive Evaluation Based On AHP Method

Multi-attribute comprehensive evaluation problems widely exist in society, economy, management and other fields. With the development of society, Multi-attribute comprehensive evaluation problems are getting more and more complex. It is significant to develop a system based on the existing achievements of comprehensive evaluation theory to solve some complicated comprehensive evaluation problems.The main researching work of this paper includes three parts. Firstly, a comprehensive study and analysis of the existing achievements of comprehensive evaluation theory is presented in this paper. Secondly, a general comprehensive evaluation system is developed based on AHP theory combined with Entropy method and some other indicator normalization methods. Finally, the effectiveness and correctness of the model, the arithmetic and the computer system are tested and verified by using the problem of selecting coal suppliers for power plants as an example.The research contents are as follows:1) This paper does extensive research on the existing achievements of comprehensive evaluation theory.For multi-attribute comprehensive evaluation theory, the main researching work includes comprehensive evaluation method and multi-expert evaluation method. To improve the objectivity and reliability of the evaluation result and make it easy to develop the software,as well as decrease the calculation time of the system, AHP method is chosen as the core model to solve the multi-attribute comprehensive evaluation problem, while Entropy Weight method is also used to help to solve the problem.In the aspect of multi-expert evaluation, the paper adopts the method based on AHP consistency indicators and the method based on weight vector to determine the weight of experts.What’s more, the ordering vector weighted average method is used to aggregate the judging information of experts.2) A general comprehensive evaluation system is developed based on AHP theory combined with Entropy method and some other indicator normalization methods.In order to make sure the maintainability, the system is based on the three-tier architecture of ASP.NET and B/S mode. In order to develop a system with a good user experience, the system adopts the technology of CSS、JavaScript、Ajax and so on.The system provides two evaluation models which are self-evaluation model and multi-expert evaluation model. For simple problem or some problems needing a quick decision, self-evaluation can be chosen. For complex and important problems, multi-expert evaluation can be chosen. Overall, the system has five functions, which are establishing the hierarchical model, inputing judgement matrix, inputing data, individual evaluation, multi-expert comprehensive evaluation.3)As an example, the system developed by this paper solves the problem of selecting coal suppliers for power plants.The problem of selecting coal suppliers for power plants is researched in this paper, and the problem is solved by the system developed in this paper. The result of two methods of multi-expert comprehensive evaluation and the result of three different data standardization methods are analysed.