# Electric Autonomous Mobility-on-Demand: Joint Optimization of Routing and Charging Infrastructure Siting

**Authors:** Fabio Paparella, Karni Chauhan, Theo Hofman, Mauro Salazar  
**arXiv:** [2211.12363](https://arxiv.org/abs/2211.12363)  
**Date:** 2022-11-22  
**Summary page:** https://www.summarizepaper.com/en/paper/2211.12363v1/

## Key points

- Optimization of routing and charging infrastructure for Autonomous Mobility-on-Demand (AMoD) systems
- Modeling and optimization framework considering fleet activities and charging infrastructure placement
- Time-invariant model capturing routes, charging patterns, and state of charge of vehicles
- Road network represented as a digraph with iso-energy arcs for accurate energy consumption representation
- Problem formulated as a mixed-integer linear program ensuring global optimality
- Framework can be solved in less than 10 minutes for practical real-world applications
- Two case studies using real-world taxi data from Manhattan, NYC presented to validate approach
- Jointly optimizing infrastructure siting outperforms heuristic placement policies in terms of charging infrastructure prevalence and empty-mileage driven by the fleet
- Increasing number of stations beneficial only up to a certain point
- Vehicle design case study shows deploying vehicles with smaller batteries results in lower energy consumption despite requiring more trips to charging stations
- Fleets equipped with smaller batteries consume approximately 12% less energy compared to those with larger battery capacities
- Valuable insights into optimizing routing and charging infrastructure for AMoD systems provided
- Proposed modeling and optimization framework offers efficient solution applicable to real-world scenarios
- Informed choices can be made by decision-makers to enhance efficiency in future transportation systems.

## Summary

This paper discusses the optimization of routing and charging infrastructure for Autonomous Mobility-on-Demand (AMoD) systems. With the advent of vehicle autonomy, connectivity, and electric powertrains, AMoD systems are becoming more feasible. To address this issue, the authors propose a modeling and optimization framework that considers both fleet activities and charging infrastructure placement. They adopt a mesoscopic planning perspective and develop a time-invariant model that captures routes and charging patterns while explicitly considering the state of charge of vehicles. The road network is represented as a digraph with iso-energy arcs to accurately represent energy consumption. The problem is formulated as a mixed-integer linear program, ensuring global optimality. The proposed framework can be solved in less than 10 minutes, making it practical for real-world applications. To validate their approach, two case studies using real-world taxi data from Manhattan, NYC are presented. In the first case study, the authors analyze the trade-off between charging infrastructure prevalence and empty-mileage driven by the fleet. They demonstrate that jointly optimizing infrastructure siting outperforms heuristic placement policies. Additionally, they find that increasing the number of stations is beneficial only up to a certain point. The second case study focuses on vehicle design and explores the impact of battery capacity on energy consumption. Surprisingly, deploying vehicles with smaller batteries results in lower energy consumption despite requiring more trips to charging stations. Fleets equipped with smaller batteries consume approximately 12% less energy compared to those with larger battery capacities. Overall, this paper provides valuable insights into optimizing routing and charging infrastructure for AMoD systems. The proposed modeling and optimization framework offers an efficient solution that can be applied to real-world scenarios. By considering both fleet activities and infrastructure placement, decision-makers can make informed choices to enhance efficiency in future transportation systems.