Monique Stinson: Strategic Decisions in Agent-Based Transportation Models: Methods and Data

Time

-

Locations

Online event

Abstract: Transportation has significant impacts on energy, emissions, and the economy. Freight transportation is particularly impactful: it is an economic necessity but generates about one-fourth of transportation emissions. High-fidelity agent-based modeling tools allow government agencies, researchers, and industry to study the effects of the transportation system and its sensitivity to policies. However, although strategy influences numerous transportation decisions, no existing agent-based models address strategy in a holistic manner. A lack of strategy data and modeling methods contributes to this gap. With this in mind, my core thesis is that strategy unifies decisions that individual actors make, and that this consistency can be achieved in agent-based models by using a carefully constructed model framework, strategy data, and behavioral models that account for firm strategy. In this talk, I present a new, theoretical foundation for transportation forecasting that uses strategy to unify agent behavior in upstream and downstream components of the agent-based framework. Next, I develop an innovative method to generate strategy data using readily available text, leveraging recent Natural Language Processing (NLP) methods and Principal Components Analysis (PCA). My novel W2VPCA algorithm creates strategy measurement data based on quantifying differences in word use.  Furthermore, I devise a novel mathematical system named SURTLV (for Seemingly Unrelated Regression of Tobit Equations with Latent Variables) that jointly considers strategy and numerous strategic decisions. My concept is proven by modeling Logistical Sophistication and Customer Service strategies jointly with strategic fleet and distribution center decisions by Fortune 500 companies.

Biography: Monique Stinson, Ph.D., is a Technical Manager in Argonne National Laboratory’s Vehicle and Mobility Systems Section. Stinson is the lead developer of the agent-based platform CRISTAL (Collaborative, Informed, Strategic Trade Agents with Logistics) that is being deployed at Argonne. She has extensive experience in studying the impacts of transportation systems on energy consumption, emissions, mobility, and equity, and has studied transportation systems in numerous US regions including Chicago, Detroit, Atlanta, Phoenix, Wisconsin, the Twin Cities, and Central California. Her research evaluates the system-level effects of vehicle technologies (such as electrification and hydrogen fueling), connectivity and automation, e-commerce, on-demand delivery, drones, commodity flow growth, ridesharing, transit fares, tolling, and other scenarios. This research informs stakeholders including federal agencies, industry, and local governments as they make funding decisions. Stinson is a member of the American Transportation Research Institute (ATRI) Research Advisory Committee (RAC), Transportation Research Board’s (TRB) Freight Planning and Logistics Committee, and the TRB Freight Data Committee, and is a contributor to the 21st Century Truck Partnership (21CTP) Freight Operational Efficiency Tech Team. She is a former member of TRB’s Committee on Bicycle Transportation, is Handling Editor of the Transportation Research Record journal, and serves as a peer reviewer for numerous other journals. Stinson received her Ph.D. in Civil Engineering from the University of Illinois at Chicago and her B.S. and M.S. in Civil Engineering from the University of Texas at Austin.

Zoom link

Event Contact