Agent-based L2O Approach to Electric Vehicle Routing Problem with Vehicle-to-Grid Supply

Speaker: Prasant Misra

Abstract

The use of electric vehicles (EV) in the last mile is appealing from both sustainability and operational cost perspectives. In addition to the inherent cost efficiency of EVs, selling energy back to the grid during peak grid demand, is a potential source of additional revenue to a fleet operator. To achieve this, EVs must be at specific locations (discharge points) during specific points in time (peak period), even while meeting their core purpose of delivering goods to customers. This talk will consider the problem of EV routing with constraints on loading capacity; time window; vehicle-to-grid energy supply (CEVRPTW-D); which not only satisfy multiple system objectives, but also scale efficiently to large problem sizes involving hundreds of customers and discharge stations. We present a reinforcement learning based learning to optimize (or RL-L2O) approach for EV routing to overcome these challenges. Using numerical simulations on benchmark datasets, our analysis shows that RL-L2O generates solutions 24 times faster than the genetic algorithm (GA) and MILP baselines; although with approximately 20% increase in optimality gap in terms of the cost.

Bio

Prasant Misra is a Senior Scientist at Tata Consultancy Services - Research, where he works on intelligent cyber physical systems for smart mobility. His current focus is on the development of mathematical models and the application of operations research methods for the management of electric vehicles, fleets and charging infrastructure. His past research endeavor was in the design of sensing techniques and sensor informatics for energy constrained network embedded systems.
He has published over 60 peer-reviewed scientific and position papers in the fields of cyber physical systems (CPS), Internet of Things (IoT), mobile systems, and electric mobility. He has been felicitated by numerous honors and awards for his work, of which it is noteworthy to mention the MIT TR 35 - India “Top 10 Innovators under the age of 35 in India” (for the development of new technology or creative application of existing technologies to solve problems) and the TCS “Exemplary Contribution Award” (for increasing the brand value of TCS) in 2017; the ERCIM “Alain Bensoussan” and “Marie Curie” Fellowship (for scientific excellence) in 2012; and the Australian Government's AusAID “Australian Leadership Awards” Scholarship (for his potential to make substantial impact on social, economic, and development challenges of the Asia-Pacific region at a leadership level) in 2008. He is a recipient of the Best Paper Award (in the applied data science track) at ACM India CODS-COMADS 2023; and has won the Best Poster Paper Award at ACM SenSys 2016 and EWSN 2013.