An information-based decision making framework for evaluating and forecasting a project cost and completion date

In the past, construction projects have frequently exceeded their cost and schedule resulting in financial losses to the owners; currently, there are very few methods available to accurately predict an expected cost and completion date. This may be because of unforeseen outcomes that could not have been accounted for earlier and because of the lack of proper tools to forecast the cost and completion date of said projects. To overcome these difficulties, project managers need a systematic and comprehensive decision making framework in order to pursue a successful achievement of their projects' goals within cost and on time. The main objective of this research is to develop an information-based tool for evaluating and forecasting a project's cost and completion during the execution. The research focuses on the construction phase of a project and is intended for implementation by project managers. This research proposed a cost estimating model that incorporates the Multivariate Probabilistic Analysis (MPA). This model was developed to predict potential cost overrun during a project's execution and to quantify the magnitude of the expected project cost, which is occasionally altered by unknown effects resulting from project's complications and unpredictable environments. Such a cost estimating model is useful in diagnosing cost performance and monitoring the changes of the uncertainty as a project progresses. This changed amount at a consistent confidence level was computed, such that the proposed framework can be used as one of the indicators for a warning signal. Bayesian Inference introduced in this research was utilized to forecast project progress and completion date in the early stages as well as all construction stages. Using this inference, project managers can combine an initially planned project progress (growth curve) with the reported information from ongoing projects during the execution. In addition, they can dynamically revise the initial plan and quantify the change of uncertainty for the completion date. Particularly, this proposed information-based tool addresses the effects of an informed data of completed work packages on the re-estimates of incomplete work packages by the use of the MPA, while assessing the impacts of a reported progress data on the variation of the uncertainty on the forecasted completion date by the operation of Bayesian Inference. The information-based decision making framework proposed in this research was developed for effective project control in quantitative and objective assessments. This framework is unique in the sense that it is mathematically derived and because it deals with the behavior of uncertainties and its impacts on the expected project cost and completion date corresponding with actual reported data of a progressive project. Accordingly, this research offers an efficient tool to assist project managers in improving their management strategies. Finally, building projects are applied to test the proposed framework and their results are analyzed to illustrate its capabilities.