Green vehicle routing problem with mixed and simultaneous pickup and delivery, time windows and road types using self-adaptive learning particle swarm optimization
This research focuses on the third-party logistics (3PL) management in sustainable reverse logistics industry that involves fuel
consumption and emission concerns based on the comprehensive modal emission model (CMEM) in transportation operations on either
deliver finished products to customers or pick-up malfunctioned/expired products or perform both operations for recycling or waste
management at the depot. We formulated a novel mixed-integer linear programming (MILP) model for an extension of the green
vehicle routing problem with mixed and simultaneous pickup and delivery problem, time windows, and road types (G-VRPMSPDTWRT) that yields optimal solutions and proposed a self-adaptive learning particle swarm optimization (SAL-PSO) to improve the quality
of solutions in large problems. Our work aims to minimize total transportation costs, including fuel consumption costs and driver costs.
The validation of SAL-PSO was conducted by the comparison of the optimal solutions obtained from CPLEX and the best solutions
obtained from the standard and proposed meta-heuristics. The relative improvement (RI) between the standard PSO and the SAL-PSO
in the G-VRPMSPDTW-RT was 0.15-7.31%. The SAL-PSO outperformed the standard PSO by the average of 3.25%.
Keywords
Mixed and simultaneous pickup and delivery, Sustainable reverse logistics, Particle swarm optimization, Self-adaptive learning