Climate Science and Policy

Adoption of electric vehicles is surging across the state, country, and world, driven by government policies to reduce carbon emissions from the transportation sector. To maximally reduce emissions of EVs, however, drivers must charge their vehicles when clean electricity generation, such as solar and wind power, is abundant. In California, this means charging during the daytime when most people are at work.

Workplace charging plays a pivotal role in this context. Many EV drivers, especially those living in apartments or rented accommodations, lack access to home charging options. For these drivers, workplace charging provides a critical solution, enabling them to charge their vehicles during the day when renewable energy is most available. Moreover, workplace charging can significantly alleviate range anxiety, making EVs a more viable and attractive option for a broader segment of the population. Although the affordability of EVs has improved significantly, the challenge remains in finding reliable and accessible charging stations. Workplace charging addresses this issue, aligns with the goal of equitable access to charging infrastructure, promotes adoption, and supports the wider transition to electric mobility.

This study examines drivers' charging behavior at charging facilities at the University of California San Diego and is extensible to any workplace. The primary motivation is to analyze the heterogeneity in where and how EV drivers charge their vehicles. By mining natural variations in the data, the study aims to inform institutional policies and planning that encourage workplace charging and deliver a positive charging experience for drivers. 

A. Project scope and methodology

Using datasets on drivers’ preferences around charging, charging sessions, and UCSD’s EV charging network, this project conducted a detailed analysis of EV drivers’ charging behavior, focusing on both the spatial and temporal aspects of charging. The data for this study are derived from enrollment surveys of 806 real (anonymized) UCSD EV drivers, alongside more than 55,000 unique charging sessions retrieved from the two main charging service providers at UCSD—ChargePoint and PowerFlex. Key components of the study include:

Imbalances in the demand for charging and the supply of chargers across campus. Understanding the demand-supply imbalance in charging sessions across various campus locations is crucial. The study identifies garages with high demand but relatively few chargers, leading to significant disparities and underutilization of network efficiencies.

Driver preferences for campus charging location vs. what is revealed by their real charging behavior. The study compares drivers’ stated ideal campus charging location with actual charging session data to identify discrepancies and analyze supply-demand imbalances across the campus that may cause deviations in charging location.

The depth, or size, of sessions that drivers’ initiate. Analyzing whether drivers engage in deep or shallow sessions and how these behaviors are distributed spatially across campus. Frequent shallow sessions are identified as a significant factor leading to an underutilized and inefficient charging network.

Identification of driver traits that affect session depth. The study emphasizes identifying commuter traits that influence EV charging infrastructure needs at both micro and macro scales. This includes demographic factors, such as their affiliation, where they live, access to home charging, and commute distances.

The analysis of EV charging behavior at UCSD reveals critical insights into the utilization patterns, demand-supply imbalances, and session depths across different campus zones. By examining the data, several key findings have emerged that highlight the unique challenges and opportunities within the existing charging infrastructure. These findings provide a foundation for targeted improvements to enhance network efficiency, equity, and overall user satisfaction. While the study focuses on behaviors at UCSD, the lessons learned are anticipated to be generalizable to other workplaces outside the campus.

B. Key findings

Supply and demand imbalances:

There are significant disparities in the availability of charging infrastructure across different campus zones. High-demand garages such as Athena, Gilman, and Scholars experience notable supply shortages, leading drivers to frequently deviate from their preferred charging location and charge elsewhere. Six garages with high demand but few forthcoming chargers were identified – Bachman, Campus Point East, Campus Point West, Rady, School of Medicine, and Keck.

Charging session depth:

Data reveal that garages with lower demand-supply ratios have higher rates of shallow charging sessions. These shallow sessions are less efficient and can contribute to congestion at charging stations. Encouraging deeper charging sessions can improve network efficiency and parking garage utilization.

Influencing factors:

Access to home charging, commute distance, and driver demographics significantly impact charging behavior. Drivers without home charging access, particularly those from lower-income groups, face greater challenges in finding available charging spots and have higher deviation sessions. Temporal patterns show peak usage times coinciding with typical work hours, stressing the need for optimal charger placement and availability.

C. Conclusion and recommendations

This study on workplace EV charging behavior at UCSD provides crucial insights into the significant disparities and heterogeneity in the demand for charging and supply of charging infrastructure across different campus zones. The findings reveal prevalent shallow charging sessions and high demand-supply imbalances in specific garages, leading to network inefficiencies and driver inconveniences. To address these issues, the following strategies are recommended:

Prioritize installing new charging stations where supply-demand imbalances are greatest:

Prioritizing the installation of new charging stations in garages with high demand-supply imbalances, such as Athena, Gilman, and Scholars, to address current imbalances is crucial for mitigating deviation sessions and encouraging deeper sessions. It is important to focus on maximizing kWh sales, EV throughput, and charger cost recovery while also considering goals of access and equity.

Encourage deeper charging sessions:

Implementing incentives such as kWh-based pricing may encourage deeper charging sessions, reducing the frequency of shallow sessions and optimizing charger usage. Additionally, developing targeted support programs for drivers lacking home charging options, including dedicated charging slots for long-hour charging.

Optimize charging schedules:

Introducing a reservation system or time-based access to manage peak usage times, distributing the charging load more evenly and reducing congestion during peak hours may influence behavior heterogeneity.

Continuous monitoring and adaptation:

Establishing a continuous monitoring system to assess the performance and utilization of the charging infrastructure will inform decisions on future expansions and improvements. Preferably, focus on those parking garages where actions are most required. The enrollment survey-led incentive mechanism already exists as an effective method to communicate behavioral benefits directly to users.

Expand public awareness and education:

Awareness campaigns to inform drivers about the benefits of optimal charging practices and the impact of their behavior on network efficiency may increase emphasis on deeper charging sessions, including improved commute efficiency and potential incentives.

By adopting these recommendations, UCSD can enhance the performance and user satisfaction of its charging network, supporting the broader adoption of electric vehicles and contributing to sustainable transportation initiatives on campus.