CAREER: System Design of Crowd Logistics via Participatory Agent-based Modeling

University Of Texas At Arlington, Arlington TX

Investigators

Abstract

This Faculty Early Career Development Program (CAREER) grant will contribute to the advancement of national prosperity and economic welfare by yielding insights into crowd logistics platform design. Crowd logistics is a term that describes crowdsourced transport and delivery of goods and freight. Senders source cost-effective logistics services from individual carriers via an online platform, which is controlled by a digital management system. To be successful, a platform must rapidly scale up its sender and carrier networks. If there are too few participants, senders and recipients will be dissatisfied by unfilled service requests, carriers will have insufficient opportunities, and the initiative will fail. This award supports a fundamental understanding of crowd logistics platform design for robust network growth and performance by attracting and retaining participants having a variety of economic and social motivations. The knowledge gained from this research will facilitate the use of crowd logistics for resilient food distribution in the face of large-scale disruptions, such as the COVID-19 emergency. The educational plan seeks to implement participatory modeling activities that will improve STEM students’ capacity for systems thinking, which is critical in the design and analysis of complex logistics systems. This research will produce a novel agent-based modeling approach to systematically explore the impacts of a crowd logistics platform’s design on its ability to quickly achieve a critical mass of participants. This approach will model the degree of control that a platform management system assumes over participants’ digital exchanges. At one end of the spectrum is a platform characterized by centralized control, full automation, and algorithmic management; at the other end is a platform with decentralized control and community-driven exchanges. Existing research on crowdsourcing platform design suggests that the degree of centralization of control impacts platform growth over time. However, understanding and predicting network effects that will lead to long-term platform growth is challenging, because the feedback loop perpetuated by participant decisions and interactions yields complex and dynamic system behavior. This research will address this challenge by representing carriers and senders as autonomous, heterogeneous, and adaptive agents, whose decisions to participate in a crowd logistics platform impact the participation of other agents over time. Participatory modeling sessions with crowd logistics practitioners will be leveraged for model validation, to ensure that agent behaviors correspond to those of their real-world counterparts. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.