Simulating Construction Bidding Using Agent-Based Modeling

Competitive bidding is the main mechanism for allocation of construction projects and consequently price determination of the construction services in the A/E/C industry. While different aspects of construction bidding have been studied in the literature, there is still a need for developing a comprehensive model that captures the complex dynamics of bidding environment by considering interactions among its components, most importantly construction contractors. This paper discusses the advantages of agent-based modeling in simulating the construction bidding process over the previously applied methodologies. Later, a developed model is introduced and used to simulate the construction bidding environment. Finally, the model is used to compare the effectiveness of major quantitative methods, known as Friedman, Gates and Fine methods, in the bidding environment under a variety of scenarios including low to high level of uncertainty in the estimated cost. This study shows that agent-based modeling is a powerful methodology for simulating and analyzing complex construction bidding environment. In particular, the result suggest that risk attitude has considerable impact on bidding performance of contractors and moderate risk averseness is the optimal policy in a long run. Also, using Friedman model can result in higher market share whereas using Gates model can result in higher profit per project.