Developing a Predictive Model of Schedule Estimates for Pre-construction Activities of Transportation Projects

There has been increasing interest from state Departments of Transportation (DOTs) in scheduling the pre-construction activities for highway projects; this interest can be largely attributed to several advantages of accurate estimates for these tasks. Developing accurate schedule estimates of pre-construction activities plays a vital role in the success of such projects and decreases their overall costs significantly. However, a review of the literature reveals more comprehensive and advanced studies on the construction and post-construction (maintenance) phases are available compared to pre-construction activities. This gap emphasizes the need for predictive models for other phases of highway projects. Such models can also be beneficial for Metropolitan Planning Organizations (MPOs) and local government agencies by providing them with a decision support tool when assessing and prioritizing projects for funding, offering them a method to assess the feasibility of completing a project within its proposed duration. Despite the importance of such estimates, many planners within DOTs suffer from a lack of statistical data and must develop timeline predictions within the boundaries of their experience and the information they already have. The literature indicates that timelines developed in this way contain significant errors when compared to actual schedules. The purpose of this research is to provide a model which can provide more accurate estimates of the duration of pre-construction activities in highway projects. A review of the literature has resulted in categorizing all pre-construction tasks into five main activities: environmental clearance, design, right-of-way acquisition, utility adjustment and bidding & letting. Among various drivers that may impact durations of these tasks, five were selected to develop the predictive models based on their reported high significance and relevance in affecting project duration: 1) Length of project, 2) Estimated cost of project, 3) Project class, 4) Population density of project location, and 5) Project type. Aggregating these durations results in the overall duration for design and pre-construction activities. To accomplish the objective of this study, data on past transportation projects was collected using various data mining techniques and analyzed using multiple linear regression. After analyzing and providing the duration estimation model for each activity, the accuracy testing was performed on each model. Finally, conclusions and recommendations were drawn based on the assumptions, limitations and findings.