A Study On Books Purchase Duplicate Amount Optimum And Supplier Selection

With the continuous development of higher education and the prosperity of the publishing industry, book procurement in university has increased year by year. Then there is a need to use a open and transparent way to regulate the procurement act in procurement process, in order to improve the quality of procurement, reduce procurement costs, avoid embezzlement and black case work. In this context, bidding as a procurement method has been gradually adopted by university libraries. Practice has proved that, this method not only reflected the basic market criterions which require open, fair and impartial, but also made full use of procurement funding.Howerver, there are still some problems in the bidding process of book procurement. The staffs with limited professional knowledge always draw upon their past experience to determine the duplicate amount. As a result, the libraries do not receive a reasonable use of procurement funding: some kinds of books can not meet the needs of reader and some others are seldom borrowed. Besides, these libraries also lack a stringent set of evaluation criteria in vendor selection, the price factor is concerned only with the neglect of the quality of books and other factors such as arrival time. All of these result in choosing the wrong provider and greater losses. How to determine the duplicate amount more scientifically and reasonably and design a set of methods of vendor selection have become two issues need to be resolved as soon as possible in domestic libraries.To solve the problem of duplicate amount optimum, this article introduces the particle swarm optimization with death penalth function. After algorithm design, realize it with java, and do analysis of the calculation results of the simulation example in the chapter four to show the effectiveness of the method.To solve the problem of vendor selection, this paper first establishes a set of book vendor evaluation index which includes both quantitative and qualitative indicators based on previous studies, then gives the specific steps of dealing with quantitative indicators using Grey Relational Analysis, which is also used in determining index weight, after these steps, the rank of each book vendor on quantitative indicators comes about. While in qualitative aspect, this paper adopts a method of multi-attribude group decision making under pure linguistic information, every expert can fill in the questionnaire based on the language scale established by their own language preference, then what this paper has to do is making statistics of the completed questionnaires according to the calculation rules of pure linguistic information, and finally getting the rank of each book vendor.