Predictive maintenance: How to avoid costly downtime
By Mike Nicholas, Director, Manufacturing Business Programs Services. Microsoft Services on April 19, 2017
Filed under Discrete Manufacturing
Variability is the enemy of any plant manager or manufacturer. To get the upper hand, you need insight into how equipment is performing in real time to know when, and where, failures are likely to occur. Using this data-generated intelligence, you can proactively schedule maintenance to avoid unplanned and costly disruptions, either on the manufacturing floor or in your customer’s environment.
Manufacturers across industries already generate tons of data. The question is: How can you best access, collect, and analyze those millions of data sets in a meaningful way, so you can monitor and optimize your products or equipment performance and intervene effectively before failures occur?
Minimizing variability, improving quality
Digitally transforming operations through predictive maintenance lets you effectively monitor your production resources and products. The benefits are twofold. First, you can forecast maintenance needs and avoid potential failures. And second, you can maintain consistency during production and increase first-pass yield and product quality.
Predictive maintenance combines the Internet of Things (IoT), big data, and the power of the cloud with machine learning to help you create and maintain reliable, high-quality equipment. An easy way to start is by monitoring your operations in the cloud. Connecting resources or products with sensors lets you make smart use of IoT and start gathering information to understand behaviors. With those insights, you can profile machinery under normal operating conditions and visualize and identify any failure history. Finally, by adding machine learning, you can recognize patterns that allow you to anticipate exactly when to intervene before a machine fails.
With predictive maintenance in place, you can begin applying intelligence to reduce maintenance costs, plan service events, and reduce downtime in your manufacturing operations.
Data insights lead to new revenue opportunities
Imagine if you could use these predictive insights to track and identify potential failures in your products used by other manufacturers. Some manufacturers are already doing that.
One leading jet engine manufacturer now uses the power of the cloud and machine learning to help airline customers anticipate maintenance needs and avoid costly downtime and delays.
Every engine they build is equipped with many sensors, generating thousands of signals that allow the manufacturer to foresee what types of maintenance should be scheduled. The manufacturer shares these insights with customers, enabling them to proactively schedule maintenance personnel and have replacement parts ready at exactly the right place and time. This reduces airlines’ inventory costs and minimizes the disruption of unplanned maintenance activities, saving them millions of dollars and allowing the manufacturer to better serve customers. As a result, the company has transformed itself from a manufacturer selling jet engines to a propulsion service provider with new revenue potential.
At Microsoft, our digital advisory team is here to partner with you, helping you harness big data and migrate it to the cloud to improve quality and lower costs. We can show you how to best use analytics and machine learning to support maintenance activities and make predictable, cost-effective repairs.
|Visit us at
|Learn how to transform manufacturing with
|Conduct an Envision
the Art of the Possible
workshop to ignite