Data-enabled Decision-Making in Emerging Co-opetitive Transportation Markets with Ambiguity

点击翻译按钮获取中文摘要

Recent communications and technological advancements, along with a growing abundance of sensor and user data, enable new mechanisms of cooperation and matching of transportation services to demand. These newfound capabilities highlight changes in moving transportation systems from centrally controlled, publicly owned and operated infrastructures to decentralized system elements. These systems involve multiple stakeholders who operate in competitive environments wherein individuals or facilities may be linked through an underlying common market problem creating benefit from cooperation. That is, these environments are co-opetitive. The overarching objective of this research project is to develop and test concepts that frame dynamic, uncertain and decentralized transportation problems in newly arising co-opetitive environments and create efficient, data-enabled optimization and equilibrium algorithms to support the operation of these future transport services. Developed methods will aid in operating peer-to-peer systems, where individuals can become suppliers, as well as system aggregators. The models and solution methods will exploit personalized information to customize services to individual system users and to better understand transportation systems of the future. Thus, this work will enable assessment of novel market concepts and mechanisms for regulating them. Through the development of an on-line library of e-learning snippets in the form of blog posts, vignettes and video clips, and efforts to involve underrepresented persons and first-generation college students, this project will broaden participation of underrepresented groups in research and positively impact engineering education.Concepts from game theory, probability modeling and ambiguity, data analytics, and model predictive control will be employed and extended for use in real-time multi-player, multi-level, ambiguous settings with high dimensionality. Closed-form solutions will be accompanied by rigorous mathematical derivation and proof. Solution methods that can be applied both for long-term decision horizons and real-time implementations will be created. Both will take advantage of information through learning about consumer/user behavior and that of competitors. This work aims to support transformation of the field of transportation, bringing concepts from diverse areas of economics, mathematics and control to envision and run markets and collaborations of the future, while simultaneously providing lasting value through fundamental contributions to both modeling and solution methodology.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.