Virtually every Digital Transformation program has Reliability and Smart Maintenance as a key focus, given it is typically the primary area with the largest opportunity to eliminate waste, reduce unplanned downtime and create a safer work environment.
But if so, why have so few manufacturers implemented even the most basic Condition Based Maintenance (CBM) program vs ‘old school’ operator rounds and reactive maintenance strategies?
Well, there are untold numbers of reasons why people and organizations continue to do what they have always done – but the fact of the matter is that much of the building blocks for intelligent maintenance already is possible on equipment today.
By simply capturing existing data from equipment at the edge – along with some basic analytics, businesses can quickly unleash valuable insights – and begin their transformation from “fix and replace” to “predict and prevent” today!
Studies have shown that predictive maintenance may reduce maintenance costs by up to 40% and dropping equipment capital investment by 3 to 5 percent by simply extending the useful life of machinery in manufacturing, providing a huge economic impact. (Source: McKinsey).
Unfortunately traditional time based (preventative) maintenance is only appropriate for 18% of industrial assets and 82% fail based on random patterns.
In fact, the act of preventative maintenance is a leading cause of unplanned equipment failure. (Source: arcweb)
Every asset in a plant will tell you before it fails, but are you listening?
Although most equipment deemed ‘critical’ is already integrated into existing automation systems with 24/7 operators reacting to real-time alarms, up to 80% of assets are considered ‘stranded assets’ like secondary pumps, motors, tanks, conveyors, robots, etc.
Most are likely integrated into a local controller but often their ‘non-critical’ status means they typically are destined to run to failure due to lack of resources.
These ‘essential’ assets are usually the primary blind spot of a factory/plan and are responsible for the majority of unplanned downtime – as they indirectly impact the overall business.
Routine based maintenance and roaming technicians with clipboards and RFID readers (or often more sophisticated vibration or leak detection systems) may be able to detect some issues before a failure. However, they cannot be everywhere at once.
True Condition Based Maintenance ensures that you track problems before they become severe, providing time to be proactive vs reactive.
Early warning indicators drastically increase efficiencies including proper maintenance priority triaging, avoiding last minute rush orders for expensive parts and overall planning to optimize business operations – ultimately reducing unplanned equipment downtime by up to 50%. (Source: McKinsey)
Still looking for that crystal ball to predict problems before they happen? Believe it or not, it takes very little information into order to provide actionable insights.
Although there are certainly complex IoT systems with tons of new battery powered wireless sensors designed to detect virtually any anomaly – most machines have existing data available which when properly measured/monitored can be very valuable in providing early warning indicators to impending problems.
A basic Prescriptive reliability program uses machine run time + MTBF specs from component manufactures to maintain equipment per use based schedule. In other words, you only need to know START/STOP information (easily accessed from the PLC or controller) plus some basic logic - possible from basic industrial edge devices to full enterprise asset management applications (on-prem on cloud). This data is the building blocks for an overall equipment effectiveness (OEE) program to move away from historic reactive bad habits. But why stop there?
By leveraging only a few more data points to provide additional perspectives, a basic Predictive Maintenance (PdM) strategy may be possible. Examples of early warning indicator data that may already be available in the local PLC/controller include:
Basic Machine ON/OFF, runtime
Use existing tags + Simple Math on key variables – monitoring pressure/temperature/flow/power load for bad behavior based operating out of safe range and/or anomalies. This might be possible direclty on the PLC/controller or nearby on an edge device with local logic, or elsewhere via a 3rd party monitoring application.
HMS provides a range of smart industrial edge devices which natively interface to virtually any device or PLC/controller with support for up to 50 different protocols and fieldbuses, making connectivity to OT equipment easy and secure.
Onboard data logging, monitoring and visualization allows for key variables to be stored and utilized to enable key machine health insights required for equipment reliability.
Most machine well-being problems can be anticipated by simply monitoring existing key data points as early warning indicators- both locally, across a factory – or across an enterprise.
HMS solutions include easy integration into virtually any application including MES/ERP systems, Data Historians and Enterprise Asset Management (EAM) systems and even Internet of Things (IoT) cloud platforms like Microsoft Azure and AWS IoT systems.
HMS Networks even provides engineering services to help integrate a full end to end solution – ensuring rapid time to value and avoiding pilot purgatory by owning the problem from proof of concept to deployment.
The path to Digital Transformation can be quite possible to execute if you follow a prudent walk before you run strategy – collect key data, utilize it for actionable insights (like predictive maintenance), integrate into broader systems – then use this foundation to safely move on the such lofty IoT aspirations as AI, Digital Twins, Machine Learning etc.