Machine learning for predictive maintenance

This post was originally published in spanish in March of 2019.

In the last months, I’ve been learning about Machine Learning (ML). I am fascinated by the potential of this technology to predict equipment failure. Predictive maintenance has been around for a while, and the technology available for data acquisition is not only mature but finally becoming affordable. However, I was never quite clear on the feasibility of applying it in the real world. You either had to pay astronomical sums for complex software, or spend time you didn’t have trying to extract valuable insights from statistical analysis. Nevertheless, Machine Learning shows promise, which is why I am gathering information wherever I can. Here are some interesting links:

• Aplicación de técnicas de Machine Learning con regularización al diagnóstico de fallos en motores de inducción,Final Degree Project by Carlos Del Pozo Gallego.
• Machine Learning for Predictive Maintenance: A Multiple ClassifiersApproach , one 2015 paper published in IEEE Transactions on Industrial Informatics .
• Big Data and Machine Learning for Predictive Maintenance, a lecture by Paul Peeling, from MathWorks.