Genetic algorithm for selecting and scheduling interdependent projects

Investment planning for systems with interdependent queues, such as a series of interacting waterway locks, is a very challenging problem. Under budget constraints, it must be determined which projects should be selected and when they should be funded to minimize the present value of total cost over the planning horizon. Due to interdependence among service stations (e.g., locks), investment planning for such systems cannot be analyzed separately for each station. Published models on investment planning for interdependent projects are rare and difficult to apply to a series of queuing systems. A genetic algorithm approach is presented in this paper for solving project sequencing and scheduling problems for such systems. A numerical example is designed to investigate the proposed algorithm. Results indicate that the proposed approach is quite efficient in finding very good solutions.