Date of Award
Master of Science
Department of Systems Engineering and Management
Alfred E. Thal, Jr., PhD
The prediction of construction time performance is a problem of interest to both researchers and construction industry practitioners. This research seeks to identify significant factors which may influence construction durations for Air Force Military Construction (MILCON) projects to establish a time prediction model. Data were collected for 856 MILCON projects completed between 1988 and 2004; this included both traditional facility and non-facility (e.g. airfield pavements, utilities) projects. These data were analyzed using Bromilow's time-cost (BTC) model (1969) as well as multiple linear regression. Neither model produced acceptable results for non-facility projects; however, the multiple linear regression model was found to provide the most acceptable time prediction model for facility projects. As with the BTC model and previous research reported in the literature, there was a significant correlation between cost and duration. However, several other factors were also identified that resulted in significantly lower than average construction durations. These include projects completed within certain management groupings (referred to as Major Commands in the Air Force), projects where the Northwestern Army Corps of Engineers served as the construction agent, and projects completed using in-house design services. Several possible reasons may exist for these differences; therefore, it cannot be inferred that the results are indicative of the organizations' management processes. The forecasting ability of the model was then evaluated using a set of 129 projects not used in the formulation of the model. The resulting model appears to provide a valid alternative for predicting construction durations for Air Force MILCON facility projects. Therefore, it may be used as a prediction tool or as a policy setting tool.
DTIC Accession Number
Hoffman, Greg J., "Estimating Performance Time for Air Force Military Construction Projects" (2005). Theses and Dissertations. 3793.