contactusurl /en/inthegroove/Search No Results Animation is not fully supported in Internet Explorer versions item found items found item found items found of Search & Help Showing On this page Your Browser is outdated Dealer Distributor Official Dealer Premium Partner address addresses Articles Regulatory Press Release,Regulatory press releases only EG. Food Hoses conformity evolution Webinar Selector Button
Is Required.
decodedHtml

Predictive Maintenance - A Service Ecosystem Is Emerging

Date: 31.10.18
Digitalization is causing a revolution in many aspects of our lives.
Here Professor Fromm, an honorary professor at the Karlsruhe Service Research Institute, provides his vision on how artificial intelligence can revolutionize the way manufacturers plan maintenance and optimize their production in the future.
Industrial services have been underestimated for a long time, despite the fact that they provide excellent opportunities for revenue and profit growth, customer satisfaction and competitive advantage. Industrial firms must realize that not only product, but also service innovations determine their future, and that the service area can benefit just as much from digital transformation as, for example, research and development, sales or operations.
Digitalization will change the services of industrial firms

Today, the service portfolio of industrial firms is still largely dominated by classical industrial services such as maintenance, repair, and spare part supply. Digitalization, which has significantly changed our social lives, has also reached the service sector and will change it dramatically.

Cheaper and almost unlimited computation power in the cloud, faster and ubiquitous internet connectivity, and low-cost electronics create completely new possibilities for industrial services through the Internet of Things (IoT). With the connectivity provided by IoT, Original Equipment

Manufacturers (OEMs) can communicate with the machines installed at their customers’ sites. They can – if the customers want it – keep an eye on technical parameters and intervene in machine processes, if necessary. They can offer their customers condition monitoring, remote diagnosis, remote maintenance, and teleconsulting.

Predicting when maintenance is required

A real breakthrough, however, is achieved when connectivity is combined with artificial intelligence. The most promising application of this type is predictive maintenance. Predictive maintenance should not to be confused with preventive maintenance, which is done in regular intervals based on time, usage or condition (see Figure 1). Instead, predictive maintenance builds on condition monitoring and uses intelligent algorithms to predict the impending occurrence of a machine failure when certain condition changes have been detected.

Maintenance can now be scheduled flexibly within a remaining time interval before a failure occurs. Timely proactive maintenance prevents damages and expensive machine downtime. It leads to an overall cost reduction and efficiency increase. There are testimonials that show that predictive maintenance yields cost savings of over 55 percent over preventive maintenance and nearly 70 percent over reactive maintenance. No wonder predictive maintenance is on everyone’s lips today.

Making predictive maintenance work requires engineering knowledge

Are there IT solutions for predictive maintenance on the market yet? Even if some vendors claim it, I doubt that you can buy a turnkey solution for your individual maintenance problem from any of them. It is clear that an IoT platform is needed, and on this platform, a predictive maintenance solution can be built.

But each situation requires careful consideration as to which sensors should be used for condition monitoring and what algorithms should be used for predictive analytics. Researchers have spent a lot of effort in developing such algorithms. In practice, however, their applicability still needs to be proven. Single components, such as rotational equipment, are better understood than complex systems, such as industrial robots or wind turbines. The development of a predictive maintenance solution requires substantial engineering knowledge besides a knowledge of IT and algorithms.

Is there a future for human intervention after predictive maintenance?

So the question is: who will build the predictive maintenance solution of the future? My prediction is that there will not be a single player, but that a vital ecosystem will emerge that puts predictive maintenance to work. OEMs and component manufacturers will lay the foundation to equip their products with sensors, electronics, and connectivity, and IT vendors will provide platforms for storing, processing, and analyzing the data. Smaller IT companies and component manufacturers with strong competencies in advanced analytics will develop specialized solutions for selected industrial domains – in close cooperation with equipment engineers. Over time, these solutions will become more and more standardized and broadly applicable.

Will algorithms replace the work of a service technician in the future? The answer is definitively no. Even if an algorithm will be able to predict an occurrence of a failure, analyze the reasons for failure, or give instructions on how to resolve the failure, human experience and human judgement is still needed for the maintenance task. The most intelligent algorithms will not replace human beings,but support their work and make them more efficient.

The role of the service technician will change

Other technologies like smart and wearable devices have a similar effect. But with the available technology and intelligence, the role of the service technician will change. The service technician will remain the person that knows the equipment installed at the customer’s site best and thus will become a trusted advisor for the customer and an intermediator between the manufacturer and the customer.

BIOGRAPHY

Prof. Dr. Hansjörg Fromm obtained a Diploma in Computer Science at the University of Erlangen-Nuremberg. In 1982, he was awarded a Doctor's degree for his research in the field of modeling and performance analysis of computer systems. After a research stay at IBM Thomas Watson Research Center, Yorktown Heights, New York he joined IBM Germany in 1983.

In the mid-1990s, Hansjörg Fromm forged the business area Supply Chain Management within IBM Global Business Services. In 2006, he was appointed European Head of the IBM Centers for Business Optimization (CBO).

Hansjörg Fromm has been lecturer at the University of Nuremberg-Erlangen. In 1993, he was appointed honorary professor and elected as a member of the IBM Academy of Technology (an international circle of about 200 IBM scientists) and appointed IBM Distinguished Engineer.

From July 2011 until July 2014, Prof. Fromm was director of Karlsruhe Service Research Institute (KSRI) at the Karlsruhe Institute of Technology (KIT) and head of the research group Service Innovation and Management. After his retirement from IBM, Hansjörg Fromm remains dedicated to KSRI as an honorary professor.