Rail-enabled last mile: simulating parcel flows via urban metro networks

Abstract

This study develops a discrete-event simulation (DES) of a prospective metro-enabled last-mile delivery system integrating warehouse dispatch, rail transport, and parcel locker distribution. The model represents the full parcel flow from peripheral warehouses to metro depots, train-based transfer, and station lockers using operator-provided planning-demand inputs and explicit operational assumptions. Selected operational objects are represented individually and follow rule-based logic when this is needed to reproduce routing, capacity, and availability decisions within the DES. The analysis focuses on throughput, demand satisfaction, vehicle loading, and heuristic-derived fleet requirements. Six experiments combine three monthly demand scenarios with two customer pickup policies. Each setup covers 30 daily demand inputs and is repeated 30 times with different random seeds. Reported mean outcomes show demand satisfaction ranging from 85.9% to 99.6%. Under the fixed 80-parcel locker-capacity assumption, the lower satisfaction observed with longer parcel-retention times is consistent with reduced downstream locker availability, while vehicle requirements vary only modestly. The study contributes an integrated representation of coupled warehouse–depot–metro–locker operations and demonstrates how end-to-end simulation can support resource and tactical planning for a predefined case-study configuration.