Research On Two-round Reverse Auction Mechanism For Online Centralized Procurement

With the rapid development of E-commerce,online reverse auction has become an effective procurement way for large enterprises and government departments to strength the control of procurement cost.In the last decades,more and more researchers in commerce community and operational research community have begun to concern on the design problem of new reverse auction in the complicated online centralized procurement because of the popularization of Internet and the emergence of global procurement market.The traditional one-round reverse auction often meets a series of great challenges since the purchaser cannot know in advance the willingness and ability of supplier.For instance,the combination explosion of purchasing items or the mismatch problem of demand and supply could decrease the effect of actual reverse auction.Therefore,this thesis will study and investigate several actual online centralized procurement problems by using the relevant methods in the fields of purchasing management,online auction,mathematical programming and intelligent computation based on the mechanism of systems engineering.The detailed research works can be summarized as follows.(1)This thesis presented a detailed review on the relevant research works in auction theory,online procurement and reverse auction.Through this research survey,the working principle of reverse auction in the purchasing management was further comprehended,which was very important to the following research works in this thesis.(2)The first major research work in this thesis was to investigate a two-round bundle-combination auction mechanism for multiple inseparable items.Firstly,a one-round combination auction mechanism was introduced and its difficult issue in the implement steps,that is,the negative influence of items' combination explosion upon the procurement effect,was analyzed.Secondly,a two-round bundle-combination based auction mechanism was designed and the bundle optimization problem and the winner determination problem were proposed due to their importance in this two-round auction.Finally,the detailed models and algorithms of these two optimization problems were discussed and studied and the relevant simulation experiments validated the effectiveness of the proposed models and algorithms(3)The second major research work was to investigate a two-round reverse auction mechanism for single separable items.Firstly,a one-round reverse auction mechanism was presented and its difficult issue in the implement steps,that is,the mismatch problem of demand and supply,was analyzed.Secondly,a two-round division-based reverse auction mechanism was designed and the division optimization problem and the winner determination problem,which were two important optimization problems in this way of two-round auction,were proposed.Finally,these two optimization problems were studied in detail from two aspects of building model and designing algorithm.(4)The final major research work was to investigate a two-round procurement mechanism for multiple separable items based on auction plus negotiation.Firstly,a reverse auction based procurement mechanism was presented and the winner determination problem in this traditional auction was studied;in particular,two different mathematical models was built with different objective functions of minimizing total procurement cost and maximizing total procurement save respectively,and a genetic algorithm was designed for optimizing the proposed models.Secondly,a two-round auction and negotiation based procurement mechanism was designed to deal with the difficult issue of the traditional one-round auction based procurement mechanism.Finally,three important problems,that is,winner determination,negotiation order and negotiation rule in the proposed two-round procurement way,were studied and the corresponding solution algorithms were presented.Simulation experimental results showed the proposed two-round auction and negotiation based procurement mechanism was an effective way for this procurement problem of multiple separable items.