Research On The Model Of Process-based Reverse Auction Problem By Bi-level Programming

Nowadays, as network and electronic commerce becoming enterprises sourcing trend, the requirements of enterprises in terms of cost and efficiency cannot be met by the traditional sourcing method, but by e-procurement. Reverse auction, as one of the main ways for electronic procurement, is both popular in practical application and in academic research. This study describes the reverse auction by a bi-level distributed programming, in which considers the average unit price, delay of delivery time, etc. for multi-issues, and interaction between the buyer and suppliers. Then mixed algorithm can be constructed to solve the above model, in order to minimize the buyer's average unit price, minimize the delay, optimize the buyer's overall satisfaction level, maximize the profit of suppliers, and achieve the largest system optimization goal.Based on literature investigation, this paper reviews online reverse auction. This paper presents a method for solving a sealed-bid, multi-issue, multi-sourcing reverse auction problem, where a buyer distributes his demand to multiple suppliers and each supplier responds by submitting a bid price to the buyer. The problem is formulated as a bi-level distributed programming model in which the buyer is an upper-level decision-maker, while suppliers at a lower level make decisions independently to each other. The interaction process is facilitated via iterative exchanges of decision information between the buyer and suppliers. In the mixed algorithm for the model, we used ant colony optimization algorithm to establish an optimum quantity allocation at the upper level.In the lower level decision making process, the methods of traditional customer order processing, and linear programming are employed to coordinate pricing and production scheduling decisions.Finally, the paper realizes the hybrid algorithm by Matlab7.0 simulation software, and also conducts three groups of simulation experiments to assess the quality of the proposed solution, as well as to examine its computational efficiency under various parameter settings. The results were consistent with the expectation.Due to the online reverse auction problem is an N-P difficult problem, more and more researchers are attracted in improving algorithm to solve it, so this research owns prospective and theoretical value. Through this iteration process, buyers and suppliers are pursuing to optimizing the whole system. The upper and lower decision makers can find an optimal solution and the relative importance of each issue through the learning process.