Using improved genetic algorithms to facilitate time-cost optimization

Time-cost optimization problems in construction projects are characterized by the constraints on the time and cost requirements. Such problems are difficult to solve because they do not have unique solutions. Typically, if a project is running behind the scheduled plan, one option is to compress some activities on the critical path so that the target completion time can be met. As combinatorial optimization problems, time-cost optimization problems are suitable for applying genetic algorithms (GAs). However, basic GAs may involve very large computational costs. This paper presents several improvements to basic GAs and demonstrates how these improved GAs reduce computational costs and significantly increase the efficiency in searching for optimal solutions.