|Ph.D Student||Harel Michael|
|Subject||A Game Theory Model for Two-dimensional Optimization of|
Process Flow in Construction Projects
|Department||Department of Civil and Environmental Engineering||Supervisor||Professor Rafael Sacks|
|Full Thesis text - in Hebrew|
The nature of the contractual relationship between subcontractors and project managers in construction projects leads to instability of resource allocations in projects. An apparent factor that explains the instability is lack of inherent trust in allocating resources. Subcontractors perform work in many projects simultaneously: because their resources are limited, they have to make decisions regarding the allocation of resources between projects. The goal of this research was to develop an economic behavioral model of decision making in a ‘two-dimensional’ environment with multiple projects.
The methodology included a series of expert interviews, formulation of the hypothesis, development and implementation of a two-dimensional game theory model, internal validation, and two experiments to test innovative management ideas. A preliminary field study discovered that fully 85% of subcontractors believe that less than 60% of the promised work will be made available; and 48% of project managers counter the lack of trust by exaggerating their demands by at least 20%.
To establish the model, a utility function for the subcontractor, and the weights of its parameters, were developed. The model was implemented as a simultaneous normal form game in a computer simulation program using Microsoft Excel, GAMBIT software and Visual Basic functions.
Internal validation of the final model was done by testing four propositions for sensitivity using comparative statics, in which one parameter is tested while the others are frozen. The parameters tested were: payment terms, project planning stability, contractual fines and project manager type (classified according to their attitude to fines and approval of accounts).
One of the more interesting findings from the two dimensional models was that in a stable work planning environment, the subcontractor will provide resources in a balanced form and the project manager will not exaggerate his/her demands. This behavior can lead to the conditions of a "win-win" situation.
After validating the model two experiments were conducted. They showed that the model can be used effectively to test proposed novel management and/or contractual arrangements, so that the model serves the goal that it was created for.
The main contribution of the study was deepening of the knowledge and the comprehension of the workings of the two dimensional subcontracting environment. The computerized simulation model can help in understanding the relationships between project managers and subcontractors from the viewpoint of managing production in a building project. It also enables prediction of results of operational decisions.