Improving construction contract administration utilizing multiattribute statistical analysis on bid stage information

This study focuses on basic construction engineering management techniques used in handling construction projects. Its ultimate goal is to investigate trends in the fields of contracting, contract administration and claims, to reduce the overall cost of engineering projects, and to minimize conflicts and confrontation between parties involved This work moves toward accomplishing this goal by making logical steps toward more effective evaluation of the construction management environment. This dissertation has explored different concepts of quantitative measures relating to performance of construction contracts. It was possible to create an integrated mathematical model that links statistical findings of projects' past performances with decision criteria in order to benefit future projects. This integrated model employs many different sub-models, i.e. , statistical hypotheses testing, multi-variate statistical analysis and multicriterion decision theories. This integration assists in reducing the subjectivity involved in some decision-making processes by quantifying issues impacting contract performance and assists analytically in predicting acceptable judgments. More specifically, the study allows the investigation of the determinants of construction contract performance in a quantitative manner to predict and then improve overall performance. Proceeding from this statement of intent, we developed logical quantitative approaches to deal with both the number of bidders and the owner's bid evaluation policies. Utilizing Order Statistics, we derived an analytical technique that recommends an acceptable number of bidders to be invited based on a given set of different contracting environments. The procedure provides consistent choice of number of bidders to be invited based on expressed preferences for a set of relevant decision attributes. To deal with the issue of owner's bid evaluation policy, a multi-variate regression analysis was also developed to investigate the performance of construction projects with respect to changes in a set of operational variables. The significance of each operational variable can then be verified to determine its level of importance to the success of construction contracts. Operational variables that are found to have conclusive impact on project performance were collectively integrated into an interactive multicriterion decision sub-model. The implemented concepts of multicriterion decision making and interactive methods for decision support, instead of absolutely satisfying all the constraints as done in linear programming, attempt to obtain a solution which satisfies as close as possible a set of objectives or goals. This is achieved by minimizing the linear function of the deviation variables within different priorities. These concepts of interactive methods for decision supports, such as those introduced in the literature, have the potential to contribute both conceptually and practically to improved contract performances and claims mitigation.