A Hybrid Delivery Planning Platform Considering Truck, Robot, and Crowdsourced Delivery
Date
2024-05
Authors
Aghakhani, Sina
Major Professor
Advisor
Committee Member
Journal Title
Journal ISSN
Volume Title
Publisher
Altmetrics
Abstract
The growing demand for advanced solutions in last-mile logistics is critical to reduce delivery expenses, ease traffic in urban areas, and minimize environmental pollution. In response, this paper explores a promising concept: a two-echelon truck-based robot-crowdsourced delivery system featuring en-route and parking charging for last-mile delivery in logistics operations. The focus is on addressing operational challenges faced by logistics service providers, specifically optimizing routes for vehicles transporting customer parcels from depots to selected optimal parking locations. From these locations, an optimal number of autonomous devices, such as robots, are dispatched from vehicles to perform last-mile deliveries. Additionally, the system embraces the efficiency of crowdsourced delivery, where regular people drivers contribute to the timely and efficient delivery of orders, further enhancing the adaptability and resilience of the logistics network. Utilizing the period when trucks transport robots offers the opportunity to recharge the robots, contributing to the improved efficiency of the distribution system. This paper introduces a comprehensive multi-objective optimization model designed for the cost-efficient routing of a truck-and-robot system specifically tailored for last-mile deliveries with time windows. The model aims not only to minimize overall delivery costs but also considers penalty costs for late deliveries, and optimizing the total system delivery efficiency. This innovative solution demonstrates its potential in significantly reducing costs and traffic, making it a valuable contribution to the ongoing discourse on sustainable urban logistics. To make sure our proposed model is effective and valid, we perform a sensitivity analysis using the LP-metric method. Additionally, we speed up the process of finding the model's pareto front for smaller instances by applying an augmented epsilon constraint method.
Series Number
Journal Issue
Is Version Of
Versions
Series
Academic or Administrative Unit
Type
Text
Comments
Rights Statement
CC0 1.0 Universal
Copyright
2024