Broker-based trade allocation in market-based multi-agent environments

e-marketplaces have been growing rapidly in the recent years due to the development of Internet technologies. Many bright outlooks of e-marketplaces will lead to appear the new highlight of electronic transaction markets and the trend of intelligential business environments in the future. Although e-marketplaces over buyers access to a huge number of products provided by sellers and allow sellers access to more potential buyers, there is a great need of efficient and scalable methods for buyers and sellers to interact with each other to achieve business transactions successfully. Thus, brokerbased trade allocation, where a broker works as an intermediary between multiple buyers and multiple sellers, is an attractive research direction in recent years, especially when broker-based trading processes are carried out under the complex electronic transaction environments. Multi-agent technologies are the major technologies for developing broker-based approaches in market environments. This thesis investigates the challenging issues of broker-based trade allocation in market-based multi-agent environments so that the allocation pairs between buyers and sellers are determined to satisfy buyers' requirements under the consideration of multi-attribute trading and different objectives.