Research Of Multi-Unit Combinatorial Auctions Based On Agent

Combinatorial auctions can be used to reach efficient resource and task allocations in Multi-agent systems where the items are complementary or substitutable. Due to the well expression of complementary or substitutable, combinatorial auction has attracted many attentions of researchers from economics, computer and engineering science. Determining the winners in combinatorial auction is NP-complete problem, and the winner determination in multi-unit combinatorial auctions with reserve prices is a more intractable one. To our knowledge, the research into multi-unit combinatorial auctions with reserve prices considered is more or less ignored.We have discussed how to avoid income losing by using reserve prices in multi-unit combinatorial auctions, which is due to the collusion and insufficient competition in the auctions. And we research into how to increase profits income by using reserve prices. A new algorithm for multi-unit combinatorial auctions with reserve prices is proposed, which is based on Sandholm's work. An efficient heuristic function is developed for the new algorithm. Experiments have been conducted. The experimental results show that auctioneer agent can find the optimal solution efficiently and can improve profit income with our algorithm in the multi-unit combinatorial auctions with reserve prices for limited problem scales. The result of our research provides the possibility to carry the multi-unit combinatorial auctions with reserve prices into executions in resource and task allocation fields such as multi-agent systems, E-commerce and Grid Computing.