Transforming IT
cancel
Showing results for 
Search instead for 
Did you mean: 

Don’t Just Predict the Future – Shape It with HPE Digital Prescriptive Maintenance

Services_Editor

By Pierre Baudelle, 

Consulting Strategist, HPE Pointnext WW Center of Expertise AI, Data and Emerging TechnologiesHPE20160525596_800_0_72_srgb.jpg

 We humans have put a lot of effort, down through the centuries, into understanding the future. Whether it’s predicting crop yields, anticipating market behaviors, or planning equipment lifecycles, there are crucial advantages in understanding what’s coming next. And even better than knowing what’s on the horizon is knowing why it will happen and what we should do about it, even automate the action. We’ve grown steadily more skilled at engaging the future, and now, with the help of our artificial intelligences and cognitive technologies, we’re seeing some exciting breakthroughs.

We’re seeing them particularly in the industrial context, where companies are looking for new ways to minimize costly interruptions and process breakdowns caused by equipment failures and maintenance activities. In just about every sector you can think of, from energy, transportation to manufacturing and distribution, organizations want to move from disruptive, reactive maintenance to a more proactive, future-ready approach.

From reactive to prescriptive

Companies are at different stages of this journey. Most organizations have implemented condition-based monitoring. An operational dashboard checks incoming data – such as temperature, speed, vibration, time – from sensors on the equipment, whether it’s an engine, wind turbine, machine tool, connected car, or a whole production line. If the data exceeds specified thresholds, the system triggers an alarm, and maintenance teams can intervene before problems occur. But condition monitoring provides only limited predictive capabilities, and can lead to unnecessary maintenance operations.

Predictive maintenance goes a step beyond this by introducing machine learning and advanced analytics. These artificial intelligence elements tap into data generated by IoT devices and equipment, including historical data, to provide a much more complete picture of the future states of key assets.

Predictive systems alone don’t provide much in the way of guidance for resolving issues, however. They may produce outputs such as “Component X has a 90 percent chance of failure within two weeks.” But a bare statement of probability like this, while useful, is not enough for the maintenance technician who has to decide how to respond; it may not even be enough to convince him or her that intervention is actually necessary. Even the most experienced technician will want to see more detail, more interpretation. What’s causing the problem? Is it the pressure or the vibration? How is it fixed? What are the costs involved?

Prescriptive maintenance goes beyond predictive by supplying that guidance. It leverages artificial intelligence to provide explanations and recommendations for preventive maintenance actions. Prescriptive maintenance can structure priorities, automate actions to prevent failures, and even outline probable financial impacts.

Perscriptive Maintenece blog.JPG

 The right actions at the right time

HPE Digital Prescriptive Maintenance Service, from HPE Pointnext, combines technologies from HPE and select partners in a comprehensive solution that provides a practical basis for your maintenance decision-making.

The solution analyzes a wide variety of performance data to identify future issues and provide recommendations to guide maintenance based on problems databases search. It support what-if analysis (i.e what if vibration is fixed first?) and can include detailed information on parameters such as criticality and estimated repair costs to support decision making. It provides a dashboard to help him or her explore the data, from a tablet, and establish priorities with full knowledge of the facts. In addition, HPE Digital Prescriptive Maintenance service can initiate responses automatically through actions such as scheduling maintenance, sending an e-mail to a manager, or generating a ticket. When integrated with a localization system It can also locate and notify a technician close to the machine.

It all adds up to more flexible, faster responses to potential issues, reduced human error, and lower risk. HPE Digital Prescriptive Maintenance helps you reduce downtime and delays due to equipment malfunctions on the one hand, and unnecessary check-ups and fixes on the other.

Optimizing the supply chain – and beyond

Prescriptive maintenance can improve your supply chain operations, too. When you know that a machine’s performance is going to deteriorate, when that’s likely to happen, and why, you can get the right replacement gear ready so you can intervene at the optimum moment. The analogy that comes to mind is Formula One racing, with its tight integration with supplies, strategically planned maintenance routines, and ultra-compressed response times, all in the service of maximizing performance.

We at HPE see a huge future in the future – that is, in systems and methodologies that help businesses understand and shape future outcomes. Indeed, we’re already seeing some interest in applying digital prescriptive maintenance technologies in the adjacent area of quality control. I’ve worked with companies that wanted to use a prescriptive approach to understand the correlation between certain product attributes and the probability of defects, and decide what actions to take, potentially without human intervention. Companies want to exploit data of very different origins and formats, including unstructured data (images, diagrams, texts) as well as telemetry data. In principle, prescriptive analytics could apply to anything that you would like to check all the time by processing signal data.

If you’re interested in any type of predictive maintenance use case, drop me a note in the comment box below. Most companies will want to start out with our HPE Artificial Intelligence Transformation Workshop (that is not just targeted at predictive maintenance topic but AI overall), which helps you develop a common understanding and vision around your use case and roadmap – I’ll have a lot more to say about the Workshop in my next post.

Learn more about HPE Digital Prescriptive Maintenance Service here, and check out the video below for a quick overview.

Resources and Related articles

About the Author:

 Pierre_Baudelle edited.jpgPierre Baudelle is Consulting Strategist in charge of technically or operationally feasible long-term AI/Big Data strategies and plans development for customers.

He brings more than 25 years of experience in IT including Consulting in Databases and Systems, pre-sales on Business intelligence and alliance management of HPE partners at EMEA level. He is also leading a Big Data inter-company Workgroup and is associate professor at Masters in Business and Engineering Schools.

Pierre earned an Engineer Diploma in Enterprise computing from CNAM Aix/France, a Master of Sciences in Databases and System Integration from University of Nice/France and a Professional Diploma in Marketing from the Chartered Institute of Marketing of Cambridge/UK.

0 Kudos
About the Author

Services_Editor

Events
Read for dates
HPE at 2019 Technology Events
Learn about the technology events where Hewlett Packard Enterprise will have a presence in 2019.
Read more
Read for dates
HPE Webinars - 2019
Find out about this year's live broadcasts and on-demand webinars.
Read more
View all