Pickup and delivery problem in the collaborative city courier service by using genetic algorithm and nearest distance

Abstract

One problem in collaborative pickup delivery problem (PDP) was excessive outsourced jobs. It happened in many studies on the collaborative PDP. Besides, the revenue sharing in it was unclear although important. This work aimed to propose a novel collaborative PDP model which minimizes total travel distance while maintains low outsourced jobs. It proposed several contributions. First, it prioritized internal jobs first rather than full collaborative model. Second, it proposed new revenue sharing model. It adopted cluster-first route-second and mixed pickup and delivery. It was developed by combining the genetic algorithm and nearest distance algorithm where the genetic algorithm was used in the clustering process and the nearest distance was used in the routing process. The simulation result shows that the proposed model was better than the comparing models: (1) combined K-means and genetic algorithm model (KMGA) and (2) combined simulated annealing and last-in first-out (SNLIFO) model. When the number of orders was high (300 units), the total travel distance of the proposed model was 37 percent lower than the KMGA model and 30 percent lower than the SNLIFO model. In average, the outsourcing rate of the proposed model was 70 percent lower than the previous models.

Guided Pelican Algorithm

“…In the manufacturing sector, they have been used to solve many scheduling problems, such as in flow-shop [1], job-shop [2], batch-shop [3], and so on. In the transportation sector, they also have been utilized to solve many routing problems, such as in the capacitated vehicle routing problem [4], where each vehicle has a maximum load capacity that cannot be surpassed, and in the pickup delivery problem [5] where the vehicle must tackle two activities: pickup and delivery. Metaheuristic algorithms are also implemented in many assignment problems [6], where certain jobs or tasks should be handled by several limited resources in the most efficient way, such as in the course timetabling problem [7].…”

Hybrid Pelican Komodo Algorithm

“…In the production process, optimization is widely used, such as in the flow-shop scheduling [1], batch-shop scheduling [2], assembly line balancing [3], procurement [4], and so on. In transportation and logistics, optimization is implemented in route planning [5], storage management [6], and so on. Optimization is also implemented in finance, such as in portfolio optimization [7], option pricing [8], credit risk assessment [9], bankruptcy mitigation [10], etc.…”