An Integer L-shaped Method for Dynamic Order Fulfillment in Autonomous Last-Mile Delivery with Demand Uncertainty
arXiv (Cornell University)(2022)
摘要
Given their potential to significantly lower costs and enhance flexibility in
last-mile delivery, autonomous delivery solutions like sidewalk robots and
drones have garnered increased interest. This paper addresses the dynamic order
fulfillment problem faced by a retailer who operates a fleet of low-capacity
autonomous delivery vehicles, servicing requests arriving in a stochastic
manner. These delivery requests may vary in package profiles, delivery
locations, and urgency. We adopt a rolling-horizon framework for order
fulfillment and devise a two-stage stochastic program aimed at strategically
managing existing orders while considering incoming requests that are subject
to various uncertainties. A significant challenge in deploying the envisioned
two-stage model lies in its incorporation of vehicle routing constraints, on
which exact or brute-force methods are computationally inefficient and
unsuitable for real-time operational decisions. To address this, we propose an
accelerated L-shaped algorithm, which (i) reduces the branching tree size; (ii)
substitutes exact second-stage solutions with heuristic estimations; and (iii)
adapts an alternating strategy for adding optimality cuts. This heuristic
algorithm demonstrates remarkable performance superiority over the exact
method, boasting a more than 20-fold improvement in average running time while
maintaining an average optimality gap of less than 1
solve a wide range of instances to evaluate the advantages of adopting the
stochastic model. Our findings demonstrate long-term cost savings of up to 20
when accounting for demand uncertainty in order fulfillment decisions.
Meanwhile, the derived savings could diminish as the uncertainty in order
arrivals increases.
更多查看译文
关键词
Dynamic Programming,Vehicle Routing Problem
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络
Chat Paper
正在生成论文摘要