(The) n-bilateral optimized combinatorial auction through a systematic combinatorial auction design methodology

Abstract

Combinatorial auctions have the potential to play an important role in electronic transactions since they enable bidding on combinations of items for complementary or substitution effects. As the interest in the combinatorial auction has increased, diverse combinatorial auction market types have been proposed to satisfy market needs. These combinatorial auction types have different market design characteristics. For effective combinatorial auction trading, a systematic approach for market design is necessary. However, previous studies have not followed such an approach. This study proposes a systematic framework for combinatorial auction market design based on three phases - market architecture design, trading rule design, and decision strategy design. This framework serves as a guideline for effective design and practical implementation of combinatorial auction markets. Next, we introduce a new and highly efficient dynamic combinatorial auction mechanism derived from our design methodology - the N-Bilateral Optimized Combinatorial Auction (N-BOCA). N-BOCA is a flexible iterative combinatorial auction model that offers optimized trading for multiple sellers and buyers in the ad-hoc supply chain. We design the N-BOCA system from the perspectives of architecture, trading rules, and decision strategies. Under the given N-BOCA architecture and trading rules, bidders and auctioneers have diverse decision strategies for winner determination. This needs a flexible modeling environment. Thus, we propose an optimization modeling agent for selecting optimal bids and auctioneers. The agent has the capability of rule-based model formulation for Integer Programming. Finally, we show the viability of N-BOCA through a prototype and experiments. The results indicate that N-BOCA yields much higher purchase and supply efficiencies than the 1-to-N (one-auctioneer to multi-bidders) combinatorial auction mechanism.