Maximize EV Charger Uptime with Vision AI Technology

In just 5 years, the number of EVs in the US will grow from about 3 to 27 million, and in line with this according to a recent analysis by PwC, the number of charging stations will increase at a 10X rate, from around 4 million today to an estimated 35 million by 2030.

However, the challenges of maintaining remote charging stations in the electric vehicle supply equipment (EVSE) industry are intensifying. There were approximately 800,000 failed charging attempts in the US last year, indicating a failure rate of 1 in 5. If this rate stays at 20%, by 2030 that could result in 7 million failed charging attempts.

EV charging manufacturers must act now to avoid scaling the problem as they rapidly grow their charging networks.

Remote diagnostic challenges in EVSE operations

Most EV charging stations are unsupervised, making it difficult for manufacturers to pinpoint the exact cause of issues when they arise. Identifying whether issues stem from equipment malfunctions, connectivity failures, degradation over time, or vandalism is not possible without physical inspection. Additionally, some stations may seem functional within network systems yet fail to charge vehicles or process payments—a discrepancy that could remain unnoticed for days. Consequently, service managers often must deploy technicians to the site not only to identify these issues but also to return for necessary repairs.

The business impact of requiring multiple technician visits to diagnose and repair the same station is significant, especially with a large network:

• Lengthy resolution times with a high Mean Time to Repair (MTTR).
• Increased operational costs owing to multiple onsite visits and increased truck rolls.
• Loss of revenue from charging stations being out of action.
• Low customer satisfaction (CSAT) scores, impacting brand equity.

How Bruviti's Vision AI solves this problem

By equipping the manufacturers with Vision AI, Bruviti enables a direct view of the issues affecting charging stations.

Here's how Vision AI works:

• Troubleshooting with enhanced image recognition: The manufacturer's customer charging app leverages Vision AI to facilitate issue resolution. By taking photos of their charging setup through the app, customers can have the AI diagnose any problems. The app then provides detailed troubleshooting steps, empowering users to independently address and resolve issues.
• Automated diagnosis and predictive resolution: Vision AI models are trained using extensive datasets, including service and warranty records and parts Bills of Materials (BoMs). These models can automatically identify specific charging stations, pinpoint the root causes of issues, and predict the necessary parts and steps for resolution.
• Integration into service workflows: The manufacturer's service management system generates work orders with detailed information about the station, the identified issue, and the parts required for repair, enabling quick dispatch of field service agents.
• Continuous model training and enhancement: The models used in Vision AI continuously adapt and improve, learning from new data to increase their diagnostic accuracy and operational effectiveness.

This approach is novel for several reasons:

• It empowers EVSE companies with a new 'vision' by enabling customers to contribute to the diagnostic process using their mobile devices directly.
• By implementing Vision AI, manufacturers can now 'see' and precisely diagnose issues without sending a technician onsite just to diagnose an issue, allowing manufacturers to save significant time, truck rolls and resources.

The outcomes:

• Faster issue resolution: Vision AI's rapid diagnostics significantly cut MTTR, ensuring issues are resolved quickly and efficiently, minimizing downtime.
• Cost savings: Precise diagnostics mean technicians arrive with the right parts and instructions, reducing the need for multiple truck rolls and lowering transportation and labor costs.
• Maintained revenue: By preventing long downtimes, Vision AI helps keep charging stations operational, preventing revenue loss from non-functioning units.
• Improved customer experience: Quicker and more reliable repairs lead to higher customer satisfaction scores. Enhanced service reliability boosts brand equity and customer loyalty.

Case study: How Bruviti's Vision AI helped a leading EV charging station manufacturer reduce service resolution time by 50%

Our client, one of the world's largest EV charging station manufacturers, wanted to ensure that their service operations were equipped to deliver at the next level as it continued to rapidly scale its network of charging ports. They faced significant delays in their repair processes, primarily due to the challenges in diagnosing issues without physical inspections. With Bruviti's Vision AI, the manufacturer realized that customers using the station could help solve the challenges by becoming an integral part of the diagnostic process.

Results with Bruviti Vision AI:

The implementation of Bruviti Vision AI resulted in:

• Reduced MTTR from 7 days to 2 days
• Saving thousands of truck rolls at an average cost of $250 per truck roll
• Minimized revenue leakage due to increased uptime
• Improved CSAT scores as a result of reduced downtime

Conclusion

With approximately 20% of EV charging attempts failing last year, electric vehicle supply equipment manufacturers must take immediate action to enhance their service capabilities. Bruviti's Vision AI empowers these manufacturers to scale their service models efficiently by making customers an integral part of the diagnostic process. This approach helps boost productivity and cost-effectiveness and significantly enhances customer satisfaction, preparing manufacturers for profitable growth as they scale.