Previous studies of construction contract auction bidding have mainly focused on finding factors affecting the markup and decision to bid (d2b), without considering expert weight and critical factors as input variables for estimating the size of markup needed. This study develops a 3-step Mamdani-type of Fuzzy Inference System (FIS) identifying critical factors and predicting markups for construction projects. The first step models the wights of 31 construction experts based on their characteristics (i.e., experience and academic qualifications). The second step models frequency, severity, and importance weights for each factor to find the rank and priority of the factors, revealing the critical factors to be current workload, project (contract) size, need for work, availability of labor and staff required, project owner, and duration. The third step takes the importance weight of each factor from the previous step and the contractor’s evaluation of the frequency of the factors as input variables to predict the bid markup. The model is demonstrated and tested with two actual construction projects having actual markups of 20% and 45%, and the predicted markups are found to be 26.3% and 42% respectively, which ensures reliable outcomes in assessing contractors’ bidding decisions. It is a novel model that can simultaneously identify critical factors and predict an optimal markup in assisting contractors’ d2b for the construction auctions and developing risk management plans. Future research can optimize this model by incorporating competitors’ bids and enhancing prediction accuracy to guarantee the lowest price with a reasonable profit margin.
|Journal||Engineering Applications of Artificial Intelligence|
|Early online date||2 Jun 2022|
|Publication status||Published - Aug 2022|