Abstract
Nowadays, logistic companies are dealing with the pressure of a highly competitive environment and trying to reduce operational costs. At the same time, customers are requiring not to wait so long to get their delivery and pickup demands satisfied. In this study, these two factors are considered with a-static-periodic solution strategy. The problem is modelled as Dynamic Vehicle Routing Problem with Simultaneous Pickup and Delivery (DVRPSPD) where delivery and pickup demands of customers continuously arrived in a planning horizon are satisfied at the same time and by the same vehicle. Furthermore, we proposed a solution methodology based on solving routing problems repeatedly for each time period by considering new and unvisited previous customers. A well-known heuristic, the Nearest Neighbourhood Algorithm (NNA), is used to solve VRPSPDs in each time period. Important real-life aspects such as demand fluctuations, different number of customers, and routing periods are integrated into the problem. According to results, the average waiting times per customer significantly increase whereas average travel times per visit notably decrease when the length of time period increases. If routes are constructed 8 times in a day, each customer waits an average 5 minutes compared to 66 minutes if it is made 4 times in the day. More routing plans cause more than 7-minute average travel times per visit. Different test settings are analysed and the results are explained to help decision makers find the best solution for both companies and customers.
Published Version (
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