Robust optimization of dynamic route planning in same‐day delivery networks with one‐time observation of new demand

Abstract

Local delivery networks expect drivers to make deliveries to and/or pickups from customers using the shortest routes in order to minimize costs, delivery time, and environmental impact. However, in real-world applications, it is often the case that not all customers are known when planning the initial delivery route. Instead, additional customers become known while the driver is making deliveries or pickups. Before serving the new demand requests, the vehicle will return to the depot for restocking. In other words, there exists a precedence relation in the delivery route to visit the depot before delivering new orders. The uncertainty in new customer locations can lead to expensive rerouting of the tour, as drivers revisit previous neighborhoods to serve the new customers. We address this issue by constructing the delivery route with the knowledge that additional customers will appear, using historical demand patterns to guide our predictions for the uncertainty. We model this network delivery problem as a precedence-constrained asymmetric traveling salesman problem using mixed-integer optimization. Experimental results show that the proposed robust optimization approach provides an effective delivery route under the uncertainty of customer demands. KEYWORDS demand uncertainty, dynamic vehicle routing, precedence-constrained traveling salesman problem, robust optimization, same-day delivery network Networks. 2019;73:434-452.wileyonlinelibrary.com/journal/net FIGURE 1 (A) Ill-placed route planning under uncertainty-the tour will back-track to a previously visited neighborhood to reach the newly emerged nodes; (B) well-placed route planning under uncertainty-the tour does not need to return to previously visited neighborhoods to reach the newly emerged nodes. Note that the white unshaded circles are the known nodes, on which the original TSP is calculated. At a particular time, the tour reaches w * , uncertainty is realized, and the new, shaded nodes are added to the problem. The tour will return to the depot, d, before visiting the new nodes

Simulation-based algorithm for determining best package delivery alternatives under three criteria: Time, cost and sustainability

“…The researchers concluded that their best policy minimized the average cost of an a priori policy by 9.1% and maximized the orders delivered. A real problem that exists for companies attempting to break into the SDD category is the ability to evolve as an e-commerce platform without compromising revenue (Ulmer, 2017;Kawa et al, 2018;Scherr et al, 2018;Yao et al, 2019). Lin et al (2018) evaluated the efficiency of on-demand same-day delivery models with regards to fuel cost, transportation time, and emission cost.…”

Modelos logísticos estocásticos: una revisión de la literatura

“…En [79] proponen un SVRP no capacitado con tiempos de viaje inciertos a lo largo de los arcos para determinar la ruta en ventanas de tiempo estipuladas, para lo cual incluyen un nuevo criterio de decisión que penaliza el nivel de servicio. Por su parte [87], consideran un problema SDD de una red de distribución local con un solo vehículo, existe incertidumbre en la ubicación de los clientes, ya que estos van apareciendo mientras el vehículo cubre otra ruta, por esto se hace necesario que el vehículo vuelva al depósito para aprovisionarse antes de servir a los nuevos clientes, además considera los costos asimétricos y su objetivo es minimizar el tiempo e impacto ambiental.…”

Route and Travel Time Optimization for Delivery and Utility Services: An Industrial Viewpoint

“…České firmy ve svých propagačních materiálech běžně uvádí, že instalací sledovacího zařízení může provozovatel ušetřit až 20 % nákladů na palivo. (Zhang et al, 2019;Yao et al, 2019).…”