Using Data to Close the Reliability Gap with Alternative Fuel Vehicles

As the next wave of alternative fuel vehicles enters the market, operators are being challenged to stay on top of new technology and address fleet reliability. Properly leveraging data and AI, companies are embracing these challenges and saving money.

There has been a huge push within the transportation and logistics industry to welcome the next wave of technology: alternative fuel vehicles. Trucks that use electricity and hydrogen as power sources will be the future (think Tesla, Nikola, Daimler or Rivian), but we already see many fleets that have made the switch to Compressed Natural Gas (CNG). CNG vehicles yield many business and environmental benefits—CNG is produced both domestically and worldwide at a relatively low cost and burns cleaner than diesel fuels. As climate change continues to move to the forefront of political and business discourse, making the switch is more important now than ever.

However, for companies that operate CNG vehicles, there’s a newer need to balance the environmental benefits with the reliability risk often experienced with operating this type of machine.

CNG Sees Failure

With any new machine or technology, there emerges new sets of maintenance experiences and challenges. Specifically, there are five common causes of low power in CNG trucks that fleet operators are seeing: turbocharger failures, ignition failures, fuel component failures, mechanical failures, and charge air system component failures (according to the Natural Gas Vehicle Institute, NGVi). To put it in perspective, one of the most critical (and common) failure points is the CNG Cylinder Head, which can have cascading impacts and cause repairs that cost between $15,000 to over $50,000. While CNG vehicles are making a positive impact on the environment and decreasing fuel costs, they can create costly maintenance issues.

Leveraging Data to Identify Failures

Predictive failure analysis can be a solution for CNG fleet operators. Analyzing available data from critical components—both upstream and downstream of the CNG Cylinder Head—allows operators to pick up on early warning signs of impending failure. Uptake is able to correlate multiple Diagnostic Trouble Codes (DTC’s) that are being transmitted by the OEM (in addition to available signals such as internal temperatures) to generate predictive insights that allow fleet operators and maintenance techs to proactively repair these vehicles prior to catastrophic failures.

This insight alone can mitigate consequential damage and avoid a $20k+ repair for this fleet operator. This is just one example of many different failures that can be uncovered and avoided with the use of AI and Machine Learning.

Why Predictive Failure Analysis?

A global food and beverage leader has turned to predictive analytics to help optimize their North American fleet of nearly 70,000 vehicles. “We are already a very reliable fleet but are looking for ways to move beyond traditional labor efficiency and reliability tactics and stay ahead of the curve,” shared the North American Fleet Manager. Their initial work in predictive analytics was focused on the electrical system and resulted in a 20% reduction in on-road failures, in addition to a 9% reduction in technician diagnostic time—an overwhelming success.

As this Fleet Manager systematically looked for the next opportunity, they focused on their CNG vehicles. “If we can move the needle on CNG reliability, it will give us the confidence to adopt these alternative fuel vehicles more broadly.”

Failure is Optional

Predictive Maintenance is no longer a moonshot—fleet operators are using AI technology today to increase the reliability of their vehicles, better plan maintenance, and improve the overall efficiency of their maintenance operations. Fleet operators need to be able to take advantage of new technologies with the confidence that their vehicles will still run reliably, and Uptake helps them get there.

A World that Always Works

At Uptake, we believe that the advancements brought on by the Fourth Industrial Revolution—advanced analytics, AI and machine learning—can enable workers to operate smarter. By ingesting data from sources such as vehicle fault data, raw sensor data and historical failure records, and then applying machine learning, operators can more effectively monitor vehicle health and risk in real-time. Imagine a world where users are alerted to potential failure before failure actually occurs.

To learn more about the impact that this technology can have on your fleet, contact us at