I-Corps: A Scalable Cloud-based Route Optimization Software for Efficient Aerial and Road Logistics

University Of Missouri-Columbia, Columbia MO

Investigators

Abstract

The broader impact/commercial potential of this I-Corps project is the development of a software application to generate optimized route plans for parcel delivery. The goal is to empower last-mile service providers who employ electric trucks and a mixed truck-drone fleet. The adoption of electric trucks and autonomous drones for last-mile logistics has risen steadily in the United States due to growing parcel volumes, rising costs, changing customer expectations, and increasing corporate climate pledges. Lack of capabilities in the existing technology limits the safe, seamless and efficient use of such emerging ground and aerial vehicles. The proposed software application may meet the emerging market needs by enabling logistic managers, route planners, dispatchers, and truck drivers with faster planning, greater visibility, real-time tracking and turn-by-turn navigation. This may allow service providers to lower supply chain costs and be competitive while facilitating last-mile delivery solutions that lead to lower carbon emissions and faster fulfillment. These capabilities may accelerate the adoption of electric trucks and drones for package delivery and alleviate the growing strain on this distribution system. This I-Corps project is based on the development of a cloud-based route optimization software that will generate last-mile distribution plans for electric trucks and hybrid truck-drone systems. The proposed machine learning and optimization-based decomposition algorithms are designed to exploit problem-specific characteristics, such as battery constraints, charging operations, and payload capacity, to efficiently solve complex routing problems. The algorithm also may account for real-life spatial (e.g., no-fly zones), temporal (e.g., time-of-day operating restrictions) and logistical (e.g., customer availability) constraints to ensure practical route plans. In addition, the proposed technology also may allow an active traffic management strategy by generating an alternative route plan in real-time using a deep reinforcement learning-based dynamic rerouting model. The proposed route optimization software may lead to new theories and contribute to the state-of-the-art knowledge on route planning methods and advance the capabilities of route optimization software to handle new logistics technologies for efficient ground and aerial last-mile delivery service. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.