Optimize fleet management with predictive maintenance

Optimize fleet management with predictive maintenance

For much of its history, the transportation sector has relied on calendars or conditions to determine when and whether maintenance was necessary. The problem with this approach is that, as we know, machines do not break down on a set schedule.

It is not surprising to see that nearly 80 percent of maintenance managers are unhappy with their protocols. They are tired of wasting money on unnecessary checkups, and they no longer want to experience the costly downtime caused by unexpected breakdowns.

Sensor data analytics for connected machines, which helps accurately predict the exact time when maintenance is required, is undeniably a complex process. However, the consequences of a reactionary approach are far scarier.

In Germany, for instance, about 25 percent of trains are behind schedule. This causes significant discomfort to passengers and freight customers. Moreover, such delays in already tight schedules have a domino effect that rattles the entire system. Reports indicate that more than 33 percent of initial delays are triggered by unexpected vehicle faults. The best part is that vehicle manufacturers can identify vehicle failure patterns in different geographies and prioritize service and component availability by region.

Unfortunately, most transportation companies have been hesitant to adopt predictive maintenance for two reasons. To start, the approach yields significant amounts of data that needs to be cleaned, organized, and leveraged within a timely fashion. Further, there is a pronounced lack of qualified data scientists in the labor market who could carry out such activities.

Luckily, a solution exists that can help maintenance managers overcome these hurdles and embrace the next frontier of fleet management.

Cognitive predictive maintenance utilizes a combination of connected technology and artificial intelligence. Data is collected from key components of a vehicle — its engine, tires, hydraulics, and more — and is then fed into algorithms that serve as virtual data scientists that reveal trends and abnormalities in real time.

Time is money in the transportation industry, and in my experience, cognitive predictive maintenance can increase fleet availability by up to 35 percent in the aviation sector and reduce material costs by 15 percent in the rail sector.

Yet those are hardly the only benefits. There are three additional fleet management perks made possible by cognitive predictive maintenance:

Reduced unplanned outages: Everyone from logistics providers to cab companies understands how critical it is to ensure consistent uptimes. GE estimates that the rail industry alone loses upward of $400 million per year because of maintenance failures and their resulting downtime.

Predictive maintenance makes it possible to identify and service a degraded component before it actually fails. Not only does that cut maintenance costs, but it also saves huge sums by avoiding lost revenue.

Improved fuel efficiency: Reining in fuel costs is one of the most persistent challenges of fleet management, and cognitive predictive maintenance allows companies to address this challenge with precision. For example, University of Nebraska-Lincoln research revealed that constantly monitoring and optimizing one single commercial truck's tire pressure could save $2,400 in fuel consumption per year.

If there is a way to optimize consumption, the data will reveal it. Additionally, if there is a way to improve engineering designs, the next generation of the fleet could be significantly more efficient than the last.

Enhanced mobility: Cognitive predictive maintenance makes it possible to plan on-the-road repairs that are intended to correct issues before they turn into failures. This leads to cost savings and eliminates much of the need for a centralized maintenance hub. Fleet managers no longer have to rely on a massively expensive facility and can instead transition to agile mobile maintenance teams.

Cognitive predictive maintenance allows transportation assets to operate at peak efficiency and productivity while also improving safety and satisfaction for customers.

Undeniably, it is a complex process, but to reject its potential would be shortsighted.

Sundeep Sanghavi is the co-founder of DataRPM, a Progress company, which tackles the business problem of maintenance inefficiencies in industrial IoT.