Using an adaptive genetic algorithm to improve construction finance decisions

Optimizing both qualitative and quantitative factors is a key challenge in solving construction finance decisions. The semi-structured nature of construction finance optimization problems precludes conventional optimization techniques. With a desire to improve the performance of the canonical genetic algorithm (CGA) which is characterized by static crossover and mutation probability, and to provide contractors with a profit-risk trade-off curve and cash flow prediction, an adaptive genetic algorithm (AGA) model is developed. Ten projects being undertaken by a major construction firm in Hong Kong were used as case studies to evaluate the performance of the genetic algorithm (GA). The results of case study reveal that the AGA outperformed the CGA both in terms of its quality of solutions and the computational time required for a certain level of accuracy. The results also indicate that there is a potential for using the GA for modelling financial decisions should both quantitative and qualitative factors be optimized simultaneously. © 2001 Blackwell Science Ltd.