Crowdsourced delivery for last-mile distribution: An agent-based modelling and simulation approach

Crowdsourced delivery is considered a possible solution to the last-mile on-demand delivery challenge. The purpose of this paper is to investigate the performance of the crowdsourced last-mile delivery with regard to service level and assets utilization. An agent-based simulation model is developed and two parameters are studied: (1) supply/demand ratio, the ratio between the number of crowd couriers delivering packages and the number of packages and (2) maximum detour time accepted by crowd couriers. The main findings are twofold. Firstly, high supply/demand ratio enables the crowd to achieve high service level with very little time spent on detour. Secondly, high levels of maximum detour time accepted by the crowd can increase service level when the supply/demand ratio is comparatively low, but it may also create competition among crowd couriers for delivery orders.