Predicting Disaster

Autosphere » Fleet » Predicting Disaster

What can telematics tell you about imminent engine failure and other costly repairs and incidents?

Telematics is breaking new ground in predicting risk, as well as important fleet maintenance issues.

Using diagnostic trouble codes to diagnose engine failures is the holy grail of telematics, according to Don Woods, Director of Client Information Systems at ARI. “We’re trying to make correlations between vehicle events and the type of wear and tear that can lead to additional maintenance costs. I don’t think anyone in the industry or the OEMs are there yet.”

ARI has employed data scientists and is working with local universities to try and “crack this nut,” but Woods says it’s very difficult. “There’s something like 2,700 diagnostic codes we’re getting on average, trying to see the frequency, and then create a model based on engine failures.

“We have a model now that we’re testing. We’re able to predict engine failures in certain cases, but the model isn’t refined enough yet. There’s too much noise in the model, but that’s where everyone would like to be, including the OEMs. We’d love to be able to say, ‘these trouble codes lead to X failure later on so, therefore, do some preventive maintenance to get it serviced.”

Predicting risk

Kimberly Clark, Telematics Product Leader at Element Fleet Management, says “We can help fleets get smarter about where their true risk is. Already, we can see lane departure events and other near misses. That kind of partial autonomous feedback helps better predict risk through traditional methods, such as looking at accident history.”

The data will enhance the risk view of the drivers to help fleets make decisions, and Clark believes there’s another opportunity around accidents themselves.

“Accidents will continue to happen, so how do we make that a more efficient process for the driver?” she asks. “If we know the location, we know what their speed was, and if they were wearing their seatbelt. We start to fill in the details.”

Overlap in experiences

By marrying maintenance programs to events coming from telematics, an experience can be created for drivers, making it easier to take action. “We want drivers to take action faster, especially with certain severe events,” says Clark. “A proactive approach is critical.”

That may involve bringing together the service fleet plan, the route and schedule of how customer activities are being handled, together with maintenance activities. “Creating an overlap in those experiences can drive productivity for the driver,” says Clark.

Predictive analytics provide data about events coming out of the vehicles to better predict component failure. “When should you be changing your tires?” asks Clark. “That can be based on not just your data, but also similar benchmarks across many different fleets of similar models and usage types.”

She believes the time is close when trouble codes can be used to predict what type of work will be needed on a particular vehicle down the road.

“Fuel and maintenance are some of the most significant costs per fleet,” Clark notes. “Much of it is correcting the process versus having a driver that has a check engine light but keeps trying to complete their day and ends up on the side of the road. There’s a lot of waste.”

Categories : Fleet

Popular Posts