Mitsubishi Fuso is interested in finding new innovative ways to incorporate AI technology including predictive maintenance software in commercial vehicles. Please get in touch with us if you have innovative solutions!

Vehicles are becoming a lot more than what they used to be. People use them for work, personal transport, and even sleep. But they now also generate tremendous amounts of data.

As soon as the engine starts, your vehicle is generating and analyzes data about its operation, like fuel usage and estimating when an oil change is needed. It is also capturing information about the outside world, including driving speed, route information, the vehicle’s load, and more.

What’s more, there is an increasing amount of internet of things IoT-connected devices and sensors in new vehicles. Most have more than 70 sensors throughout, with anywhere between 15 and 30 in the engine alone.

More than half of all vehicles sold in 2022 were connected, according to ABI Research. That number is expected to skyrocket to 70% in the next six years.

New vehicles have more sensors and cameras than a typical smartphone. Their cameras alone will generate 20 to 40 megabits per second, while radar will generate anywhere between 10 to 100 kilobits per second.

According to Intel CEO Brian Krzanich, in the future, the average car will generate 4 terabytes of data per hour of driving.

The automotive industry is increasingly becoming data-driven. The onus is on manufacturers to find ways to utilize this multi-dimensional data to elevate their products.

Why not utilize this data to turn the analysis further inward with AI?

A vehicle can check its own operations and diagnose component problems before they cause failures.

Gathering and analyzing data enables vehicle owners to move from corrective to predictive maintenance.

What is predictive maintenance?

Predictive maintenance refers to the use of data-driven methods based on AI and machine learning to detect irregularities and defects of in-service equipment.

Software leverages the data from the multitude of in-vehicle sensors as well as historical service records. Then it analyzes that information with machine-learning AI techniques to identify the health of individual components. If any issues are discovered, the vehicle systems notify the user that maintenance is required.

According to Heavy.AI, “Predictive maintenance technologies include nondestructive testing methods such as acoustic, corona detection, infrared, oil analysis, sound level measurements, vibration analysis, and thermal imaging predictive maintenance, which measure and gather operations and equipment real-time data via wireless sensor networks.”

The goal is to provide a longer lifespan for the product, greater safety, and convenience to the user while minimizing downtime.

This has even wider implications for businesses.

Imagine how predictive maintenance could benefit a truck fleet. An operator could replace or repair equipment before failure, avoiding costly vehicle off-road (VOR) scenarios that could prevent a truck from performing its income-generating activities. More importantly, it would keep the drivers and communities safer by reducing breakdowns and accidents.

Traditional methods of fleet maintenance have typically relied on time-based schedules, depending on the manufacturer and local regulations. Commercial trucks, for example, are usually inspected every 16,000 – 32,000 km.

This is known as preventive maintenance.

Predictive maintenance vs preventive maintenance

You can think of preventive maintenance as a manual approach that requires mechanics to monitor and analyze equipment in its current state. They rely on historical data, life expectancy, and averages to predict when maintenance will be necessary.

The problem with this approach is they cannot take into account real-time operating conditions. This could lead to earlier-than-necessary maintenance, or even worse, a breakdown taking the equipment out of operation.

The future of maintenance

The implications of this technology are wide. It could affect spare-part manufacturing supply and distribution, supply-chain logistics through predictive needs reporting, as well as recycling management through component end-of-life predictions.

Predictive maintenance will become increasingly important in the automotive industry as more and more vehicles become electrified. Predictive maintenance can be used to analyze battery degradation and prevent problems before they occur.

In the next 10 years, predictive maintenance will likely replace preventive maintenance across most machine-dependent industries. The benefits are simply too cost-saving and beneficial to ignore.

The industry is expected to be valued at over $28 billion by 2026, growing alongside relevant technologies like AI, digitization, and even the industrial metaverse.

Real-time analytics and data are necessary advancements for the next phase of the industry that will focus on optimization and sustainability.

You can expect your vehicle to tell you what it needs in the very near future.

At Mitsubishi Fuso, we are interested in learning how users can best benefit from predictive maintenance and AI solutions.

Should traditional preventive maintenance techniques be incorporated with predictive solutions? Can each benefit the other and form an optimal maintenance strategy?

What most interests you about predictive maintenance? Drop us a comment below!