Abstract:
Objective Aiming at the low proportion of feasible solutions and convergence difficulties in existing optimization algorithms for ship engine room layout optimization problems, this study conducts research on multi-objective intelligent design methods to achieve the intelligent layout design of the engine room.
Methods A two-stage multi-objective optimization method is proposed. In the first stage, the order of the equipment layout is used as the variable and the initial layout scheme is screened by solving the integer programming problem based on the non-dominated sorting genetic algorithm-II (NSGA-II) algorithm and mixed packing algorithm. The mixed packing algorithm integrates the ideas of shelf and skyline algorithms, with optimization objectives including space utilization rate, aisle and maintenance space, and maintenance efficiency, and constraints covering equipment interference, maintenance accessibility, exclusivity, and center of gravity, among others. In the second stage, based on the initial scheme, the best layout is optimized with equipment spacing and aisle width as variables.
Results Applying this method to the optimization of equipment layout in a local area of a ship's engine room, maintenance efficiency is increased by 17.18%, the maximum width of aisles and maintenance space is optimized by 0.47%, and the overall space utilization rate is significantly increased by 33.36%, with all optimization objectives not lower than the manual layout schemes. At the same time, parameter experiments verify the rationality of the NSGA-II algorithm parameters, elite strategies, grid parameters, and generality of the method.
Conclusions The two-stage optimization method is feasible and applicable, effectively improving the efficiency and effectiveness of engine room layout optimization, and providing a solution for intelligent layout design.