Keywords: construction claims negotiation, multi-agent systems, MASCOT model, ZEUS agent developing toolkit.

Claims negotiation plays an important role in construction claims management because of its specific functions in facilitating claims settlement, keeping a harmonious relationship between project participants, and preventing and resolving construction disputes. However, industry practice shows that claim negotiation is one of the most time and energy consuming activities in claims management. Construction claims negotiations are normally conducted in an inefficient way due to the problems associated with claims negotiation and management, such as the time-consuming negotiation preparation and negotiation process, the conflicting roles of the engineer role in the negotiations, and the complex human factors.

The development of multi-agent systems provides an innovative approach for resolving the above problems. The important characteristics of agents: autonomy, collaboration and co-operation, goal orientation and learning, make it possible to build a system in which agents, on behalf of their owners, can settle most of the quantifiable claim negotiation items in an effective and efficient way. Thus, their owners can focus on the major claim items.

The research has developed a multi-agent system for construction claims negotiation (MASCOT). Described in this paper, four key stages have been undertaken:

• first, all the major characteristics of construction claims negotiation have been analysed, the key problems have been addressed;
• three negotiation theories and their applications in multi-agent systems have been explored to select proper theoretical models for the MASCOT;
• thirdly, the MASCOT model has been developed based on the above analyses. Of particular interest is the integration of Zeuthen's bargaining model with a Bayesian learning mechanism, which addresses the characteristics of construction claims negotiation; and
• finally, the prototype of MASCOT model has been developed.

This paper describes aspects of the above four stages but focus specifically on the computer implementation of prototype system by using the ZEUS agent development toolkit. The significance of this research lies in the fact that it not only solves the inefficiency problem of construction claims negotiations, but also provides a systematic research methodology for the further application of multi-agent systems in the construction industry.

purchase the full-text of this paper (price £20)

go to the previous paper
go to the next paper
return to the table of contents
return to the book description
purchase this book (price £122 +P&P)