Shops everywhere are struggling to contain runaway repair costs. Equipment, parts and labor shortages have made managing fleet maintenance more difficult than ever.
Most fleets have extended their trade cycles to keep enough trucks on the road. Fortunately, new technology offers a way to proactively manage the maintenance of older vehicles.
Modern tools can accurately predict maintenance needs. Fleets with a modern toolbox can simplify the process of analyzing work order data and gain predictive insights to schedule repairs in advance to maximize uptime and prevent costly breakdowns and over-the-road repairs.
Cost Pressures Driving Maintenance
In trucking, almost everything can be measured as a cost per mile. And when assets are not moving, the costs add up quickly. Downtime costs up to an estimated $0.89 per minute^. Add in the average expense of a roadside breakdown and the number jumps to $1.48. The total comes up to $1,241 per truck for just one day of not generating revenue. This price tag doesn’t include late delivery penalties, the impact on driver turnover, or damage to the customer relationship.
When quantified like this, it’s no wonder maintenance shops are stressed.
Purchasing new trucks lessens the burden, but OEMs are fighting against shortages of semiconductors and basic materials like steel, aluminum, and plastic resins. As a result, truck manufacturers are pushing out delivery dates, building trucks with missing components, or canceling orders altogether.
The average cost to maintain a vehicle aged four to six years is 2.75 times higher than a vehicle less than three years old. The average age of Class 8 trucks on the road today exceeds seven years. This contributes to the 20.4% increase in maintenance costs over the last 24 months tracked by the Technology and Maintenance Council (TMC) of the American Trucking Associations.
Older assets naturally are more prone to fail. Every incident has the potential to put drivers at risk and create customer service issues that ripple throughout the company.
The Data Science Solution
So much of the maintenance that occurs on older equipment is reactive. A lack of sensor technology and telematics leaves many maintenance operations “driving blind.” The inability to anticipate repair needs means unidentified minor repairs can create major equipment failures. Paired with limited parts availability and shop backlogs, the standard two-day repair now can take around two weeks.
Calculate that downtime cost. Painful, right?
Despite running older assets without the latest technology, maintenance operations have an ample supply of data to drive their decision-making. Even the most manual operations have information capable of saving thousands of dollars and downtime hours at their fingertips. The key is analyzing the data and making it actionable.
That’s where data science for predictive maintenance comes in. Using a digital technique called natural language processing (NLP), machines can extract data from work orders. The technology then aggregates the information to show trends and statistics. History provides insight into future problems. Rather than reacting to component malfunctions as they occur, maintenance teams can get in front of issues.
For example, past work order data may uncover a consistent trend in a gasket failure across a particular model of trucks. Not only does the maintenance team learn what part is failing and on which units, they can identify when the component typically malfunctions. They then can make the repair proactively at the optimal time to maximize the lifespan of the part before creating a breakdown.
Replacing a gasket is far less expensive than rebuilding a blown engine.
Without predictive maintenance analytics, identifying this issue is extremely difficult. Most shops do not have the resources for dedicated maintenance analysts to crunch the numbers and find the trends. Even if they do, identifying the issue can take weeks of work. By that time, the analysis is outdated.
Predictive analytics can process every applicable piece of data over an unlimited period in just a matter of seconds and update the findings with every new work order.
The Value of Forecasting Failures
Having insight into how older equipment is operating using existing data is a gamechanger. The move levels the playing field between new and aged equipment, giving fleets more options to better use their capital. For example, teams can better pick which units to sustain while knowing what trucks are best to buy based on past repairs.
Maintenance teams know what vehicles need attention before they break down. Equally as important is identifying underperforming equipment, especially when it comes to fuel consumption with prices north of $5 per gallon.
The TMC cited technician skills as a top-three contributor to truck repair delays. Using predictive maintenance to identify impending issues gives technicians more time to train on the repair. The analytics also reveal technician skills gaps requiring intervention based on how their work held up over the road.
Aged trucks do not limit teams to an outdated approach to maintenance. Work order data and a predictive analytics platform can help any shop get their equipment operating like new.
Outperform Expectations with Uptake
Uptake Fleet eliminates the guesswork from maintenance management. Using data models trained through more than a billion hours of analysis, Uptake finds maintenance issues so fleets can focus on fixing them before they become failures. Users get the insight they need to prioritize maintenance decisions that drive business value. Uptake helps fleets reduce fuel and maintenance costs, increase uptime, and bring new efficiencies to maintenance facilities.
Interested in how you can plug in work order data to supercharge your maintenance? More than 200,000 assets are already benefiting from predictive insights through Uptake. Request more info on Uptake Fleet and discover how our predictive maintenance technology can transform your operation.
^Calculated based on FMCSA rules mandating a 14-hour on-duty limit for commercial truck drivers
