This research presents a relative entropy method for improving agent-based multi-attribute negotiation efficiency (REAMNE) of collaborative working in construction projects (CWCP). This method aggregates preference information of negotiators of CWCP in two steps. Firstly, compromise group preference order is ascertained by using preference information, which is newly provided by negotiators, and compromise preference information, which is formed in previously agent-based automatically negotiation process and calculated by using a compromise preference model. Secondly, group preferences are aggregated by using a relative entropy model, which is established based on entropy theory meanwhile considering the multiple attributes in negotiation of CWCP. The method of REAMNE has the merit of fulfilling the necessary requirement of group decision making (GDM), i.e. maximizing preference consistency of GDM and keeping minimum gap of negotiators' utility and group preference.
|Name||Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)|
|Conference||4th International Conference on Cooperative Design, Visualization, and Engineering, CDVE 2007|
|Period||16/09/07 → 20/09/07|
- Collaborative working
- Relative entropy
- Theoretical Computer Science
- Computer Science(all)