A Machine Learning Decision-Support System for Selecting Optimal Innovative Project Delivery Methods for Bundled Transportation Projects
CAIT project no.: CAIT-UTC-REG68
Fiscal Year: 2021/2022
Status: In Progress
Principal investigator(s): Rayan H. Assaad, Ph.D.
Performing organization(s): New Jersey Institute of Technology
Managing organization: Rutgers CAIT
In cooperation with: New Jersey Department of Transportation
Partner project manager: Nisharg Dalwadi, Project Management Specialist
Supported by: USDOT-OST-R
UTC, grant, or agreement no.: 69A3551847102
One of the earliest and most essential decisions that must be made in any infrastructure project is the selection of which project delivery method (PDM) to use. Since the choice of the PDM will dictate the language in the contract and the timing of its signing; state DOTs, local agencies, and tribal governments must make such decision early on during the pre-planning phase of the project. The primary goal of this proposal is to develop a data-driven decision-support system to help state departments of transportation (DOTs), local agencies, and tribal governments select the best project delivery method (PDM) for each bundled contract by leveraging machine learning algorithms, while also taking into consideration the specific goals of each bundle.
The intended outcome of the project is to assist DOTs, local agencies, and tribal governments in utilizing alternative, innovative contracting methods in the development of project bundling projects, programs, and initiatives to reduce costs and streamline design, construction, and maintenance activities. It is expected that the research outcomes will present a structured approach to assist agencies in making project delivery decisions, assist agencies in determining if there is a dominant or optimal choice of a delivery method, and provide documentation of the selection decision.