Press

​Maintenance software: misconceptions on predictive maintenance

Release date: 2/4/2019

Paul Moore from International Mining spoke to the main mining software players about where software fits into digitisation and optimisation, and uncovered some common misperceptions about predictive maintenance along the way.


DINGO says that after working with mines from around the world for over 25 years, it has heard quite a few concerns – and misperceptions – about predictive maintenance.  DINGO’s asset health software, Trakka® 4.5, captures predictive health information automatically from each mining machine, reports on performance, and if problems are found, schedules a service request in advance to prevent equipment failures. It maximises the data the mine already has – including preventive maintenance data – so an issue can be addressed before it becomes a problem. 

The company argues that implementing predictive maintenance isn’t predicated on doing traditional preventive maintenance well – or even at all. “DINGO’s goal is to shoulder all the heavy lifting and maximise your current processes, so your team can focus on its tasks while we implement predictive maintenance for you.

Routine maintenance is usually triggered by breakdowns and fixed factors like time, age, service recommendations, and meter readings. The downside to preventative maintenance is that it’s easy to miss something if it occurs outside the scheduled maintenance window. Conversely, predictive maintenance is based on the actual operating condition of your equipment so it’s continually assessing if everything is functioning well.” 

Some mines rely on sensor data to understand which equipment needs attention. The downside is that sensor data is only part of the story – according to Dingo’s maintenance experts, over 80% of all problems are found via other condition monitoring sources. 

Another challenge is that the amount of data gathered is often overwhelming and nearly impossible to analyse manually. On average, less than 1% of available data in the mining industry is being used – if that data was being intelligently utilised, it could prevent costly and time-consuming equipment breakdowns.  

“DINGO’s Trakka 4.5 ingests, curates, and analyses data from almost any source while recommending actions to remediate issues. By using predictive analysis and machine learning, Trakka can ‘learn’ from patterns and make intelligent predictions based on the data.”

DINGO also believes that data analysis alone isn’t enough – human expertise also needs to be applied to troubleshoot and diagnose issues. Its team of Condition Intelligence experts reviews condition monitoring data daily to proactively identify issues and recommend corrective actions. They will also continue to monitor open issues until equipment returns to a normal operating state. 

The company also argues that most mines don’t realise just how well they could perform if they had the right technology in place. Nearly every mine has room for improvement: based on data compiled from more than 50 mining operations across the globe, 33% of major components are regularly operating in a warning state and more than 11% are running in critical condition.


“Inside each of your machines is a wealth of information. Predictive maintenance and asset health software is about listening, searching, locating, and acting to fix impending issues before they become major problems.”  The cited typical payback with DINGO is greater than 4 to 1 within 12 months, says the company. 

For more information on Dingo's industry leading predictive maintenance solutions, email us at info@dingo.com or use our contact form


Paul Moore │ International Mining │ Excerpted from the feature article – Mining Software:  Software at mining's leading edge