Machines whisper before they scream: We have developed an AI model that predicts expensive problems

In most industries, maintenance is a waiting game. Things get fixed when they break. But in the 21st century, an age dominated by data and automation, this approach no longer makes sense. The solution could be predictive maintenance. It’s an approach that uses sensors and software to analyze device performance in real time and predict when it might fail.

Edward Khomotso Nkadimeng, lecturer and researcher in artificial intelligence and data systems in nuclear/particle physics at Stellenbosch University, has researched how a predictive maintenance model can help keep critical systems running – from research equipment to national infrastructure. He explains why this approach could be a practical tool for resilience across Africa.

What is a predictive maintenance model and why did you develop one?

Decades after the global industrial boom, many industries relied on a simple rule: Wait until a machine breaks, then fix it. This made sense when machines were simpler…