Research On Modeling Of Incentive Algorithm And Development Of Incentive Decision Support System With A Case Study Of Construction Conflict

The implements of construction projects are relative to many subjects, such as owners, contractors, supervisors and so on. It's difficult to reach a complete agreement among all subjects in interest, so the arising of construction conflicts is hard to avoid. As the concepts of Human-Oriented Society and Harmonious Society become more and more popular in China, it is required to explore more ways to settle conflicts effectively, and how to use scientific methods to achieve a consensus of all subjects becomes a hot issue in the field of conflict resolution.Game theory is a reasonable method for analyzing conflicts. Especially, Metagame and conflict analysis are proved to have advantages in construction conflict resolutions. However, there is a need for in-depth study in the following points. 1) The results often offer equilibria without Pareto-optimality or of weak stability, which may puzzle players and analyzers in decision-making. 2) Because of the heaviness of hand calculations, players are considered short-sighted, which makes the results be of insufficient credibility. 3) Due to the lack of computer-aided calculations, results can't be quickly obtained under conditions that feasible scenarios are excessive and there is uncertainty in players' sights and preferences.Aimed at recovering the limitations of game theory above, incentive is presented in this paper. Incentive is defined as the phenomenon that players take strategies which are favorable in far-sights but unfavorable in short-sights. Based on the concept of Incentive, algorithm improving traditional conflict analysis is built to achieve equilibria with high stability. During the building process of incentive algorithm, assumptions and basic factors involved are discussed and defined.Secondly, factors impacting scenario stability are analyzed under the framework of incentive algorithm. The reason for breaking out circles is explained, and the reasonableness of ignoring the simultaneous sanctions between players is also represented. By taking the bounded rationality theory, scenario stability is studied with different lengths of players' sights, and it is qualitatively discussed how the stability of incentive point impacts the stability of the scenario to be verified. Furthermore, incentive algorithm and conflict analysis are compared in dealing with changes of feasible scenarios and players' preferences.Thirdly, the decision support system is developed to support the incentive algorithm in computer-aided calculations. In the program developing process, C language is adopted to build the framework, and JavaScript is introduced to complete the man-computer communication, which is convenient for the operation of users.Finally, the incentive algorithm and the corresponding decision support system are tested by being applied in a real construction conflict case. This case is studied by conflict analysis and incentive algorithm respectively, and the equilibria results of two algorithms are analyzed comparatively.