Resources

Insights

An Introduction to Predictive Maintenance

May 14, 2021 | Insights

Mining is equipment intensive; the equipment is expensive to maintain, operate and replace.  In some cases, an unplanned failure can result in up to $3,000/hour of lost production or over $70,000/day. This cost alone is a significant reason why waiting for assets to fail is not an option in the mining industry.

So how do you reduce the chance of unplanned equipment failure, reduce operating cost, extend component life, and increase availability, all while conducting the minimal amount of maintenance? Sound too good to be true?

On average, DINGO customers see a 20% increase in components operating in normal conditions, improved availability and a >15% gain in component life. At the core of these results is predictive maintenance, but before we discuss PdM, lets’ look at some alternative methods.

Reactive Maintenance vs. Predictive Maintenance

Reactive maintenance is a strategy that repairs or replaces equipment after a failure has occurred.  The benefit of reactive maintenance is maximum production output up until the point of failure.  Because no maintenance was completed before the failure, the machinery was 100% operational and had zero downtime. Reactive maintenance also doesn’t require any planning or upfront costs like predictive maintenance.  But is this an effective strategy in the mining industry?

There are many problems with this maintenance strategy. For starters, the cost of an unplanned failure is likely to outweigh any added production value of running the machine to failure.  When unexpected failures occur, production targets are missed, trust is broken with customers, and premiums are paid to make emergency repairs.

In addition to cost, reactive maintenance will reduce asset health and life, and unplanned breakdowns can create an unsafe environment for workers and technicians.

Would you destroy your car’s engine because you didn’t have time for an oil change? Or put your life at risk by running your vehicle’s breaks to the point of failure? In most cases, people get scheduled oil changes and brake repair jobs to avoid critical damage, which is an example of preventative maintenance.

Preventative Maintenance vs. Predictive Maintenance

We have discussed some of the issues with preventive maintenance in our last article, “Turning Mining Data into Actionable Intelligence.” Preventative maintenance depends on fixed factors such as manufacturer service recommendations, but machines don’t wait for scheduled maintenance to fail.

Successful mining operations are proactive and anticipate failures before they happen; this is possible with predictive maintenance.

What is Predictive Maintenance?

Predictive maintenance is an effective maintenance strategy that forecasts machine failure and helps identify underlying problems that need to be fixed.

Earlier, we used an example of getting an oil change for your vehicle as preventative maintenance.  But changing the oil on a set schedule may result in oil changes too frequently–or too late–as oil wear will vary depending on driving conditions.

Many modern vehicles have sensors that constantly monitor the oil life and make a recommendation based on the oil’s actual condition rather than a set number of miles driven. Similarly, predictive maintenance is based on the equipment’s real operating condition.

How does Predictive Maintenance work?

DINGO’s Asset health software, TRAKKA®, captures predictive health information from each of your machines and then uses a proprietary analytics platform to assess equipment condition.

If problems are found, the system, which seamlessly integrates with your existing workflow, alerts the team and recommends corrective actions.

This early warning system equips maintenance teams with the actionable knowledge required to promptly rectify impending issues resulting in increased availability and fewer unplanned failures.

By continuously improving your equipment’s health and performance, DINGO’s solutions increase availability, extend component life, and reduce operating costs.

Using Predictive Maintenance in Mining Industry (Example)

One of our clients had a 150K dozer engine that was on the path to catastrophic failure after only 4,000 hours of operation; however, the lab ratings pointed to a steady-state of operation and weren’t catching the underlying issues.

DINGO’s Trakka predictive analytics engine identified and correlated several contaminants that were at critically high levels and deduced combustion issues in the engine.

As a result, DINGO’s condition intelligence experts immediately issued a work order with the steps to pinpoint the root cause of the combustion issue – these checks identified that four exhaust valves and an injector were not within OEM specifications.  The problem was rectified, combustion returned to normal, and the dozer was cleared for operation.

TRAKKA identified an otherwise undetectable combustion issue before the damage occurred, while its workflow management system enabled the maintenance team to find and fix the root cause quickly. As a result, the engine rebuild cost was avoided by the mine.

Contact DINGO for a brief Asset Health Consultation or use our Savings Calculator to estimate the annual savings that your operation could achieve with DINGO’s industry-leading predictive maintenance solutions.

businessman

“Trakka allows our team to run a highly effective condition-based maintenances program that helps our mine operate at the lowest cost per hour.”

– Maintenance Manager at a leading North American coal mine

More Resources Just for You

Power of Asset Health & Performance Dashboards

Power of Asset Health & Performance Dashboards

In order to gauge the ongoing performance of your predictive maintenance program and improve compliance, it's important to measure results that tie directly to your KPIs. Using asset health and performance dashboards, TRAKKA shows condition data from the operation...

read more
The Practical Application of Predictive Analytics

The Practical Application of Predictive Analytics

DINGO’s Trakka Predictive Analytics solution utilizes artificial intelligence and machine learning to predict impending equipment failures with confidence, allowing miners to proactively perform corrective maintenance actions to minimize downtime and optimize asset...

read more