Table 2 Input parameters of MSC-PSO algorithm.
Parameter name | Description | Typical range/Example |
|---|---|---|
Inertia weight | Dynamically adjusted weight balancing global and local search | 0.4 to 0.9 |
Acceleration coefficients | Weights for individual and social experience in velocity update | 1.5 to 2.0 |
Swarm size | Number of particles for parallel search | 30 to 500 |
Max iterations | Termination condition balancing computation and precision | 50 to 200 |
Dimensions | Number of variables to optimize (e.g., control or inventory parameters) | 3 or higher |
Chaotic mapping | Generates uniform initial solutions to avoid particle clustering | 0.5 |
Pareto threshold | Fitness tolerance for selecting non-dominated solutions | 0.1 to 0.5 |
Annealing temperature | Controls suboptimal solution acceptance with initial temperature and decay | Initial 100, decay 0.95 |
Collaborative weight | Dynamic coupling weight between inventory and routing planning | 0.3 to 1.0 |
