Online Program

A Data-Driven Approach to Project Cost Estimation for the Mississippi Department of Transportation (305288)

David Allen, Mississippi State University Institute for Systems Engineering Research 
*Jonathan Storey, Mississippi State University Institute for Systems Engineering Research 
James Warren, Mississippi State University Institute for Systems Engineering Research 

Keywords: cost modeling, statistical learning, python, GUI

For any project, an accurate estimate of the final cost is essential for determining the budget. The Mississippi Department of Transportation (MDOT) has tasked a team of researchers at Mississippi State University’s Institute for Systems Engineering Research to develop a tool that allows a user to develop accurate cost estimates based on historical project data. The tool consists of a graphical user interface (GUI) which allows a user to produce a cost estimate through the selection of different parameters and a collection of models trained on historical bid data. The modeling process consists of pre-processing the raw data through an importing, cleaning, and augmenting process then training, testing, and selecting different machine learning models. The tool constructed for MDOT will allow decision makers to update and maintain accurate cost estimates. In this presentation, we provide an overview of the project thus far, information on the developed GUI, and specific information pertaining to the developed models.