High-fairness task assignment method based on ternary spatiotemporal crowdsourcing

Abstract

A high-fairness task assignment method, e.g. for delivery tasks, is based on ternary spatiotemporal crowdsourcing. A crowdsourcing worker, a crowdsourcing task and a crowdsourcing place are defined. A path planning algorithm is used to find a delivery path among the tasks and places using a point-based representation for the delivery order. For each node, a delivery object, an earliest arrival time for the worker, and a latest arrival time for the worker are stored. Remuneration obtained by the worker for the delivery is calculated, alongside a travel expense. Data is preprocessed by finding a set of <task, place> matching pairs with a shortest total distance. An optional worker set is selected for each task based on a task constraint and an optional task set is determined for each worker based on a worker constraint. Finally, a game algorithm is used to implement task assignment. The task assignment aims to improve fairness among workers without affecting maximization of overall remuneration.

Images