Water Pump Scheduling System Using Scatter Search, Tabu Search and Neural Networks—The Case of Bouregreg Water System in Morocco

Mathematical based decision support for operative scheduling in water supply systems is highly desirable but leads very often to very difficult optimization problems. In this paper we propose a nonlinear programming approach that yields practically satisfactory operating schedules in acceptable computing time even for large system with several pumps. Based on a carefully designed software using Scatter Search, Tabu and Neural Networks in hybrid optimization algorithms, this approach employs a special initialization strategy for convergence acceleration, special minimum up and down time constraints together with pump aggregation to handle switching decisions, techniques for further speed-up. Results for selected application scenarios at Bouregreg water production system in Morocco demonstrate the high performance of the approach.