A Hybrid Particle Swarm Optimization Algorithm for Multi-Objective Pickup and Delivery Problem with Time Windows

Abstract

This paper studies the multi-objective pickup and delivery problem with time windows (PDPTW), in which a fleet of homogeneous vehicles with the same capacities located in a depot serve a collection of given transportation requests. Each request is composed of a pickup location, a delivery location and a given load. The PDPTW is to determine a vehicle scheduling strategy with the objectives of minimizing the number of vehicles utilized, the total travel distances and the total waiting times. A mixed integer programming model is built to formulate this multiobjective PDPTW. Then a novel hybrid particle swarm optimization (HPSO) is proposed to solve this problem. This algorithm adds particles neighbor information to diversify the particle swarm and use the variable neighborhood search (VNS) to enhance the convergence speed. Finally, some numerical experiments based on existing benchmark instances are given to show the effectiveness and feasibility of the algorithm.

A Robust Pickup and Delivery Problem with Uncertain Travel Time

“…For example, a tabu search combined with simulated annealing was introduced in [8] to solve the PDPTWPD. To address the same variant, researchers have used a hybrid Particle Swarm Optimization Algorithm [9]. Furthermore, two metaheuristic approaches were used to solve the SPDPTWPD: the first one is based on a hybrid genetic algorithm [10] and the second one combines a simulated annealing with a local search procedure [11].…”

“…To confirm that each route begins and finishes at the depot, we used constraints (5) and (6). The respect of each vehicle capacity is guaranteed by (7),(8), (9) and (10). Constraints (11) update the starting service time at each node for each scenario while taking into account the associated traveling time.…”

Particle swarm optimization for vehicle routing problem with fleet heterogeneous and simultaneous collection and delivery

“…In [12] one finds a proposal to solve the routing problem with homogeneous fleet of vehicles, time windows and pickup and delivery, using Cloud Particles. The proposed method works with the particles separated by neighborhoods and adds the information to the neighborhood particle swarm diversify and increase the speed of convergence of the algorithm.…”

A multi-objective memetic algorithm based on locality-sensitive hashing for one-to-many-to-one dynamic pickup-and-delivery problem

“…Multi-objective optimization methods such as multi-objective evolutionary algorithms (MOEAs) [11,15,23,34,44,49,53] are greatly desirable because of their capability of obtaining multiple trade-off solutions. Many MOEAs have been proposed for static VRPs/PDPs [18,20,21,24,29,52,66] and a few for general dynamic VRPs [22,55]. Yet, there is very limited work on multi-objective DPDP.…”