MCDM'07 - paper no. 2
Multiobjective combinatorial auctions
Petr Fiala
Abstract:
Auctions are important market mechanisms for the allocation of goods and services. Combinatorial auctions are those auctions in which bidders can place bids on combinations of items. The advantage of combinatorial auctions is that the bidder can more fully express his preferences. This is particular important when items are complements. Allowing bidders more fully to express preferences often leads to improved economic efficiency and greater auction revenues. A typical combinatorial auction problem is the so called winner determination problem. The problem illustrates the possibility to formulate combinatorial auctions as mathematical programming problems as well as the complexity of combinatorial auctions. Auctions with complex bid structures are called multiobjective auctions, since they address multiple objectives in the negotiation space. Multiobjective optimization can be helpful for detailed analysis of combinatorial auctions. Buyers can specify weights and aspiration levels that express their desired values on the attributes of the items to be purchased. Interactive methods for multiobjective optimization are proposed for analysis of combinatorial auctions and for negotiation process.
Keywords:
combinatorial auctions, preference elicitation, multiobjective optimization, negotiation, interactive methods, Dynamic Network Process
Reference index:
Petr Fiala, (2008), Multiobjective combinatorial auctions, Multiple Criteria Decision Making (3), pp. 21-34
download