Manufacturing Tomorrow


How IoT Can Improve After-Sale Performance

Jan 31, 2017
Manufacturing and Distribution


Forbes defines the Internet of Things (IoT) simply as “…the concept of basically connecting any device with an on and off switch to the Internet (and/or to each other).” Some common examples of IoT in use are wearable devices that connect your health metrics to your iPad and, on the other end of the spectrum, airplane engines connected to the manufacturer’s computer systems. General Electric predicts that in just the next few years, 50 billion assets (machine parts, sensors, monitors, appliances, etc.) will be connected in similar ways. Why? So we can measure, improve and predict performance.

A New Type of Preventive Maintenance

We’ve already outlined some of the waysIoT will improve processes by tracking assets, monitoring equipment, ensuring quality and improving safety, but there’s even more that can be accomplished using connective technology—like better understanding after-sale maintenance requirements and using data to understand and predict when a product will require service.

Let’s say you make calendars (equipment used to smooth paper under high pressure) for the paper industry. Here’s a look at what the maintenance process looks like using the traditional model, and what a method made possible with IoT looks like:

Traditional maintenance model

A fault occurs in some part of the calendar and, after a worker becomes aware of it, maintenance is contacted and thoroughly inspects the equipment. Once the nature of the problem is determined and documented, maintenance identifies what’s believed to be the most effective remedy and the person/people best able (or available) to implement the solution. The necessary parts and tools are identified and collected, and the person performing the repair proceeds to address the problem based on repair documents that may or may not be current/accurate. Once the fix is completed, the problem and solution are documented in multiple places.

IoT maintenance model

With IoT sensors and other devices in place within the calendar system, process deviations and machine failures are detected through real-time monitoring and steps to resolution are initiated automatically. The cause of the failure is fully communicated by sensors and the appropriate solution identified. The person with the right expertise is automatically assigned to implement the repair, with the right parts or tools itemized and tagged for pickup. All best-practice knowledge about the repair is in the job ticket, readily available to the maintenance person.

IoT makes it possible to move beyond relative simple failure warnings, though, to actually predict failure, essentially turning the old break-then-fix and/or every-6-months-maintenance models into a single proactive predict-and-prevent model.

Using the scenario above, predictive failure uses data being generated by the calendar every minute of every day - data about vibration, temperature, oil usage, pressure, etc. - to predict when some part of the equipment will fail. Imagine the time and cost savings this implies for customers. Think, too, about them not having to deal with expense and downtime of an equipment failure and how you can use the data to further improve your products’ design.

 Machines That Learn and Predict

IoT represents opportunities the manufacturing world is just beginning to fully appreciate. One such opportunity is in leveraging solutions like Cortana, a Microsoft digital assistant, as part of a framework for intelligent equipment systems, with each machine being capable of “learning” normal and abnormal behaviors. With that insight they’ll communicate to both upstream and downstream equipment, signaling when processes need to change to account for actions taking place.

The Harvard Business Review sees huge opportunity as the machines become capable of learning and predicting outcomes. “Machine-learning algorithms that adapt through experience and evolve in intelligence as they’re exposed to data are driving changes in businesses that would have been impossible to imagine just five years ago.” This means, essentially, that improvements in processes and performance will be driven more and more by machines themselves, not by people analyzing the data that comes from them.

Implications for Service and Revenue

What does all this mean for discrete manufacturers? While you may not be quite ready for machine-learning algorithms that re-engineer your production processes, you may be ready for predictive maintenance that helps extend the life and enhance the performance of your products - and that could generate additional income in the way of guaranteed up-time arrangements that your customers pay additional for. This concept, of creating a revenue stream around your ability to monitor and better maintain the products you make, is one that’s becoming more common as customers - especially those purchasing capital equipment - can’t afford downtime and repairs if they want to stay competitive.

It’s not too early to consider your options when it comes to predictive maintenance and how IoT could take your manufacturing operation to the next level. Why not reach out to one of our manufacturing experts? We’ll demonstrate what’s possible!


Wipfli logo square

Wipfli Editorial Team

Manufacturing Tomorrow blog
Subscribe to Manufacturing Tomorrow