Comparing the Impact of Learning in Bidding Decision-Making Processes using Algorithmic Game Theory


Although previous research efforts have developed models to assist contractors in different bidding decisions, there is a lack of research work that investigates the impact of integrating learning algorithms into the construction bidding decision-making process. As such, this paper develops a simulation framework to determine the bid decision that would result in the optimal outcomes in the long run. To this end, the authors used a research methodology based on an algorithmic game theory approach. First, data was collected for 982 US public construction projects. Second, a framework was formulated to represent the bidding decision-making process. Third, a comparison between three learning algorithms was performed, including the multiplicative weights, the exponential weights, and the Roth-Erev. Fourth, two bidding strategies were simulated: the first strategy aims to win more projects while the second strategy aims to reduce the cases the contractor might fall prey to negative profits (known as the winner's curse). The outcomes of this study demonstrated that integrating learning into construction bidding decision-making process (1) gives contractors competitive advantage over their competitors by either doubling their chance of winning more projects or reducing losses in the long run, and (2) benefits owners by ending-up paying less for their projects in the long run. Ultimately, this study adds to the body of knowledge by equipping contractors with a practical bidding framework that can be used in their bidding decision-making process to overcome the inherent complexities and uncertainties in the competitive construction bidding environment.


Civil, Architectural and Environmental Engineering

Second Department

Computer Science

International Standard Serial Number (ISSN)


Document Type

Article - Journal

Document Version


File Type





© 2021 American Society of Civil Engineers (ASCE), All rights reserved.

Publication Date

01 Jan 2021