Most companies in asset-intensive industries like utilities, renewable energy, transportation, and manufacturing are looking for a better (read: less costly, more accurate) way to optimize CapEx and manage their asset maintenance process. An ideal system would predict asset failure and help them maximize the return on their capital expenditures. It would help them answer questions such as, “What assets should I plan to repair or replace? Which assets should I continue to operate even if they shows signs of problems? Where should I allocate scarce funds so I can replace the assets most likely to fail in the future?”
Organizations, especially those with an aging asset fleet, are moving from scheduled maintenance to condition-based maintenance (CBM). Often CBM compares a statistical model for an asset class to a specific asset’s rated condition to determine which asset needs attention. That approach is leaps and bounds better than sending a crew out to a work site just because it’s the fifth year in the asset’s life. However, there is yet a better way.
New machine learning algorithms can take this paradigm shift much further. While most CBM systems attempt to predict imminent failure, they aren’t designed to reveal the probability that an asset will survive an unhealthy condition and continue operating. Superior machine learning models can ‘see over the hill’ and predict failure probability at different times in the future. The models learn and understand a specific asset’s possible paths to failure. Operators can then balance risk versus cost and provide an optimal window for replacing or repairing an asset. This maximizes the operating life of the asset, thereby optimizing CapEx. That same approach also provides the basis for optimizing the repair work, improving OpEx efficiency as well.
SpaceTime has published a white paper where you can learn how machine learning can predict asset failure and optimize asset expenditure. You can download the white paper here (registration is required).