Equity, Equality, and Need: Digital Twin Approach for Fairness-Aware Task Assignment of Heterogeneous Crowdsourced Logistics

Abstract

Industry 5.0 utilizes the Internet of Things (IoT) and autonomous computing to facilitate human–machine collaboration, where humans and machines coexist in a competitive economic ecosystem. In conventional workplaces, fairness is widely recognized as a driving force behind human motivation, loyalty, and productive collaboration. However, current fairness-aware task allocation methods have primarily focused on homogeneous workers, concentrating on either equity or equality as the sole fairness principle. With the rising trend of diverse worker fleets consisting of autonomous robots/vehicles and human-in-the-loop as service providers (e.g., crowdsourced logistics), novel approaches are necessary. Our contribution entails a fairness-aware task allocation approach for heterogeneous workers, leveraging the digital twin to understand the system’s behavior and facilitate real-time adaptation. Our proposed solution considers equity, equality, and need, utilizing the maximum-weight bipartite matching algorithm. Multiple incentive scenarios are utilized to evaluate the potential of the approach. The experimental results suggest that our multi-objective approach yields better overall fairness in various scenarios than the baselines.