Breaking News

How artificial intelligence and machine learning are benefiting fleets

“Artificial intelligence” and “machine learning” are popular buzzwords, but when used effectively can deliver tangible benefits to fleets.

Trimble examined the subject during its virtual in.sight user conference this week, inviting two fleets to share their experiences with the emerging technologies.

David Dunst, IT manager with Paper Transport, said when the company first decided to dip its toes into AI, it was to understand which drivers presented the greatest risk of being in a collision. It wanted to intervene and train those drivers to prevent such incidents.

It began by hiring a couple of consultants to help it understand the data it had collected, but Dunst acknowledged it was a time-consuming and lengthy process, taking up a lot of internal resources. For its next phase, it opted to work with Trimble, and specifically its Trimble Dispatch Advisor platform, which aims to assign the right drivers to the right loads.

But Dunst said it was important the dispatchers still had the ability to make the final decisions. The AI is designed “to provide guidance to the human, who then makes better, faster, more informed decisions,” he said. “We take the knowledge we have as humans, apply AI, and when we put those together we become so much more intelligent and better at what we do as a business.”

The other area AI contributed to improvements was in RPA (robotic process automation), or automating repetitive tasks. The benefit here was to have employees working on more important things than scanning broker sites for loads, for example.

“We are trying to continue to grow our business, and the processes we had in place where we were having to do manual work meant we had to keep having to hire, hire, hire, to keep up,” he said. “It is not sustainable if you want to grow as an organization. We took away the tasks people didn’t want to do and that opened them up to do the more meaningful, valuable work.”

Using AI to identify risky behaviors has helped the fleet reduce its significant accidents, and has improved its driver retention by creating more interaction between operations and drivers. It also uses the technology to recognize positive driver behaviors. Using Trimble Dispatch Advisor has meant drivers are being dispatched more efficiently and are getting home more frequently.

Matt Mullins, vice-president – program management with Covenant Transport, said driver retention was an area the company wanted to improve using AI. It has used the technology to improve its dispatch, while ensuring the personal relationships it has with its drivers were not sacrificed.

“We want to keep that career path conversation open, and not put it in the hands of a machine,” he said. “In our business case, it was to give us the most data we have to work with, to work very closely with those drivers to understand their needs.”

Like Paper Transport, Covenant also used it to reduce repetitive jobs such as pre-employment certifications and driver advances. Year to date, the company has saved more than 2,000 man hours, Mullins said.

(Image: iStock)

Chris Orban, vice-president – data science with Trimble Transportation, said vehicle health analysis can also be improved through machine learning. Instead of a fault-based maintenance program, fleets can anticipate when failures are likely to occur by comparing their vehicles to a massive vehicle population in similar applications monitored by the telematics provider.

Looking to future opportunities, Dunst says AI and ML will drive even bigger benefits when they’re applied industry-wide. Historically, he noted, data owners – whether they be carriers, OEMs or drivers – have been protective of their data. He wants to see that change, so Paper Transport will know when a delivery would fit more efficiently into a competitor’s network and vice-versa, or at the very least into a different division within its own company.

“If all partners come together to share knowledge, we become more efficient as a group,” he said.

If this is achieved, added Orban, AI and ML could “drive these inefficiencies we know are in our industry to 100% utilization, 100% on-time service, no driver turnover. It sounds insane to say, but we can get there.”