A neural network bid/no bid model: the case for contractors in Syria

Abstract

Despite the crucial importance of the ‘bid/no bid’ decision in the construction industry, it has been given little attention by researchers. This paper describes the development and testing of a novel bid/no bid model using the artificial neural network (ANN) technique. A back‐propagation network consisting of an input buffer with 18 input nodes, two hidden layers and one output node was developed. This model is based on the findings of a formal questionnaire through which key factors that affect the ‘bid/no bid’ decision were identified and ranked according to their importance to contractors operating in Syria. Data on 157 real‐life bidding situations in Syria were used in training. The model was tested on another 20 new projects. The model wrongly predicted the actual bid/no bid decision only in two projects (10%) of the test sample. This demonstrates a high accuracy of the proposed model and the viability of neural network as a powerful tool for modelling the bid/no bid decision‐making process. The model offers a simple and easy‐to‐use tool to help contractors consider the most influential bidding variables and to improve the consistency of the bid/no bid decision‐making process. Although the model is based on data from the Syrian construction industry, the methodology would suggest a much broader geographical applicability of the ANN technique on bid/no bid decisions.