Abstract
This paper presents a bee colony optimisation (BCO) algorithm to tackle the vehicle routing problem with time window (VRPTW). The VRPTW involves recovering an ideal set of routes for a fleet of vehicles serving a defined number of customers. The BCO algorithm is a population-based algorithm that mimics the social communication patterns of honeybees in solving problems. The performance of the BCO algorithm is dependent on its parameters, so the online (self-adaptive) parameter tuning strategy is used to improve its effectiveness and robustness. Compared with the basic BCO, the adaptive BCO performs better. Diversification is crucial to the performance of the population-based algorithm, but the initial population in the BCO algorithm is generated using a greedy heuristic, which has insufficient diversification. Therefore the ways in which the sequential insertion heuristic (SIH) for the initial population drives the population toward improved solutions are examined. Experimental comparisons indicate that the proposed adaptive BCO-SIH algorithm works well across all instances and is able to obtain 11 best results in comparison with the best-known results in the literature when tested on Solomon’s 56 VRPTW 100 customer instances. Also, a statistical test shows that there is a significant difference between the results.
Highlights
The vehicle routing problem (VRP) is the generic name given to all types of problems that involve a set of routes for a fleet of vehicles that use one or more depots to serve a geographically delimited town or set of customers
The results show that the adaptive bee colony optimisation (BCO) algorithm performs better than the BCO algorithm in terms of distance, number of vehicles, and average distance of 31 runs (Table 2)
The adaptive BCO algorithm proposed in this study for the vehicle routing problem with time window (VRPTW) problem excels in terms of both the improved quality of its solutions and shorter computational time
Summary
The vehicle routing problem (VRP) is the generic name given to all types of problems that involve a set of routes for a fleet of vehicles that use one or more depots to serve a geographically delimited town or set of customers. Its main objective is to serve all customers while imposing a limited demand on the minimum number of vehicles and/or total cost. One of the extensions of the VRP is the vehicle routing problem with time window (VRPTW), which is concerned with a predefined time interval called the time window. In this scenario, a vehicle can visit a location only in a specified time window. If a vehicle arrives at a PLOS ONE | DOI:10.1371/journal.pone.0130224 July 1, 2015
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
