Designing an Optimized Electric Vehicle Charging Station Infrastructure for Urban Area: A Case study from Indonesia

TL;DR

This paper presents a location-optimized model for EV charging stations in Indonesian urban areas, demonstrating how to maximize coverage and profit while analyzing trade-offs between cost, coverage, and service quality.

Contribution

It introduces a maximal covering location model tailored for EV infrastructure planning in Indonesia, incorporating real data and analyzing trade-offs between station number, cost, and service time.

Findings

Only four stations can cover Surabaya with current technology.

High consumer travel time (~35 minutes) impacts service quality.

Adding more stations or redundancy increases costs but improves reliability.

Abstract

The rapid development of electric vehicle (EV) technologies promises cleaner air and more efficient transportation systems, especially for polluted and congested urban areas. To capitalize on this potential, the Indonesian government has appointed PLN, its largest state-owned electricity provider, to accelerate the preparation of Indonesia's EV infrastructure. With a mission of providing reliable, accessible, and cost-effective EV charging station infrastructure throughout the country, the company is prototyping a location-optimized model to simulate how well its infrastructure design reaches customers, fulfills demands, and generates revenue. In this work, we study how PLN could maximize profit by optimally placing EV charging stations in urban areas by adopting a maximal covering location model. In our experiments, we use data from Surabaya, Indonesia, and consider the two main transportation modes for the locals to charge: electric motorcycles and electric cars. Numerical experiments show that only four charging stations are needed to cover the whole city, given the charging technology that PLN has acquired. However, consumers' time-to-travel is exceptionally high (about 35 minutes), which could lead to poor consumer service and hindrance toward EV technologies. Sensitivity analysis reveals that building more charging stations could reduce the time but comes with higher costs due to extra facility installations. Adding layers of redundancy to buffer against outages or other disruptions also incurs higher costs but could be an appealing option to design a more reliable and thriving EV infrastructure. The model can provide insights to decision-makers to devise the most reliable and cost-effective infrastructure designs to support the deployment of electric vehicles and much more advanced intelligent transportation systems in the near future.