Mastering Predictive Maintenance in the Digital Economy
Around the world, OEMs and equipment operators are investing in advanced automation technologies with the hope of reducing downtime, lowering costs, and accelerating operations. Among other things, production equipments can work fast and accurately, and they never need a coffee break.
But whether you’re using the latest automation technology or you still rely on human know-how, your operations come to a standstill when equipment goes down. As more manufacturers embrace the Digital Economy, equipment maintenance and service only becomes more important.
The good news is that technologies and techniques available today are making predictive maintenance truly achievable — for OEMs and equipment operators with the foresight to implement innovative new strategies.
Replacing Best Guess with Big Data
In the past, manufacturers managed the maintenance and service of their production lines through experience and gut feel. After mass-producing the same product over and over, you got a sense of when things might go wrong and when equipment needed to be serviced.
But as manufacturing grew more complex and distributed, gut feel was no longer enough. You needed technology to help you move from best guess to best practice, inching toward preventive maintenance that set your operations on a fixed schedule of service and repair.
Yet in today’s Digital Economy — with its extended supply chains, individualized products, and requirements for true customer centricity — preventive maintenance is no longer enough. Successful asset operators are now moving to a more mature predictive maintenance strategy.
Predictive maintenance helps reliability engineers reduce maintenance costs and unplanned downtimes by predicting failures long before they happen. Analyzing large amounts of sensor data from equipment and merging that data with business information can yield insights never before possible.
What’s more, by augmenting existing service processes with remote diagnostics, equipment manufacturer can better understand root-cause failures and improve product quality. They can also deliver differentiating and higher-margin services to their customers.
For example, one SAP client transitioned from merely making equipment to offering equipment as a service in a pay-per-use scenario. In the past, the company made and sold industrial air compressors. Today, it still makes the compressors, but its primary offering is the compressed air itself. Rather than buy the machines, its customers pay for the compressed air the machines produce.
But the only way the company could implement this innovative business model is through remote service management. This approach optimized service processes to achieve competitive advantage and more profitable service revenues.
Another SAP client, a leading provider of transportation services, dramatically improved the reliability of train operations and thereby improved customer service. Today, the company combines IoT sensors to capture 700 terabytes (TB) of data with advanced analytics to define business rules that drive predictive maintenance processes. Expected result is the reduction of its €1.3 billion annual maintenance costs by at least 8 percent.
Integrating Data for Business Results
Imagine if you could achieve similar results in your organization. Fortunately, the technology to achieve these outcomes is increasingly available.
The cost of IoT sensors has decreased dramatically over the past few years. The wireless services to connect sensors with centralized data repositories are becoming commoditized. Cloud-based infrastructures, which minimize capital expenditures and rationalize operational expenditures, are available around the world. And an in-memory computing platform that enables real-time data analytics is becoming a business requirement across industries.
Manufacturers now need to integrate these technologies into their business processes. It’s not enough to implement technologies standalone and expect results — because operational data doesn’t necessarily tell you which actions to take to achieve transformational improvements. Integrating shop-floor data with business information and processes is where you can achieve real value.
At a tactical level, you can begin to automate parts replenishment and service dispatch, dramatically reducing maintenance costs and increasing uptime. At a strategic level, you can begin to achieve business insights that lead to processes and models that help you outmaneuver the competition.
Fortunately, predictive maintenance and service needn’t be a Bing Bang approach in which you bet everything on an uncertain outcome. Start by collaborating with your CTO, your lines of business, and your maintenance operations to identify specific use cases that will generate the most business value. As you gain experience, you can move on to projects that involve more complexity but that promise even greater results.
Great read thanks 🙂