It may be said that predictive maintenance is the poster child for the “analytics of things.” Preventing down time and reducing maintenance costs has been a successful use of data science for a long time.
But sometimes there is more than just predicting a component failure or avoiding a need to replace. Consider these examples and associated case studies:
|Besides Predicting The Maintenance…||Example Case Study|
|Recommend parts or alternatives a technician should use for a given fault code, using the longest lasting, cost effective parts.||A Recommender System for Field Service Technicians|
|Suggest actions based on a machine-learning based root cause analysis||Reducing Manufacturing Scrap Through Root Cause Analytics|
|Identify and prioritize maintenance issues that pose a threat to safety.||Machine Vision for Detecting Equipment Problems|
|Zero in on maintenance calls that result in the greatest reduction of parts and labor costs||Predictive Maintenance for Medical Instruments|
|Link model predictions to the financials of false alerts to avoid the cost of unneeded down time.||Predictive Maintenance for Trucks|
|Forecast parts and labor requirements||Forecasting Maintenance Requirements|
|While on the service call, address what issue might be next||Predictive Maintenance for Medical Instruments|
|Determine which parts to stock on the truck||A Recommender System for Field Service Technicians|
|Use the same data to uncover product quality issues||Uncovering Product Quality Issues using Machine Data|
In many instances the data used to predict maintenance issues can be used for examples like these. Your analytics strategy should consider the various direct and associated use cases related to maintenance.