Throughput-fairness tradeoffs in mobile task fulfillment platforms

Abstract

This thesis studies the problem of allocating tasks from different customers to vehicles in mobility platforms, which are used for applications like food and package delivery, ridesharing, and urban monitoring. A mobility platform should allocate tasks to vehicles and schedule their order to optimize both throughput as well as fairness of the achieved task completion rates across customers. We find that there are interesting, new tradeoffs between fairness and throughput caused by shared mobility, and develop a method using guided optimization to manage the tradeoffs to achieve both high throughput and fairness across customers. We evaluate our system, Mobius, on traces gathered from ridesharing and aerial sensing applications.