The evolution of maintenance
The science of maintenance is on the cusp of a transformation. The Internet of Things, combined with advancements in edge computing and analytics, is poised to usher in an era of what is referred to as “prescriptive maintenance”. But for many companies’ operations and maintenance teams, daily maintenance tasks feel like a grind. It’s critical for firms to understand the maturity curve of maintenance, so they can determine:
- Where their operations currently are
- Where they want to be – that is, where will they get the most return for their investments in technology and processes
- How to evolve their maintenance programs.
Let’s examine how maintenance has evolved over the years, and where it’s headed in 2017 and beyond.
The old method: time-based maintenance
For many companies, maintenance has been conducted in the same manner for decades – it’s based on manufacturers’ recommendations. Nearly every valuable piece of equipment comes with a set of recommendation on how to maintain the equipment based on insights from an engineering or R&D team that created the product. Often compliance with time-based maintenance requirements is required as part of leasing or warranty terms.
A good example of time-based maintenance is car leasing. You agree to rent a car and drive below a mileage threshold. You are covered by a warranty for the duration of the lease so long as you maintain the vehicle based on manufacturer’s recommendations, which typically require service at either time or mileage intervals – whichever comes first. Of course, if your leased car breaks between planned maintenance events, you simply address the issues as they come…but this is rare with new leased vehicles.
This is an outdated and inefficient way to conduct maintenance, when technology enables far more sophisticated and cost-effective methods and products themselves have become far more reliable due to superior engineering.
The most widely accepted current method: condition-based maintenance
With the advent of small-scale computing technology, such as embedded sensors in valuable equipment, companies can now engage in condition-based maintenance. Instead of maintaining equipment based on a pre-defined schedule, this type of maintenance looks at an asset’s actual condition to determine the need for maintenance.
Before sensors were cheap and ubiquitous, condition-based maintenance was often done by veteran maintenance teams. This might involve physical inspections of critical equipment or rely on simpler technologies for determining condition (ex: a pressure meter reading).
With the automation of many industries and the explosion of computers and sensors, condition-based maintenance has become machine-led. Sensors built into equipment provide real-time readings to centralized systems, that help maintenance teams maintain equipment before problems occur. Technologies such as edge computing are making it easier to follow condition-based maintenance programs in environments where connectivity is a problem, such as on moving assets (ex: a ship) or remote assets (ex: oil rigs).
In our experience, most companies have either adopted or are working towards implementing rigorous condition-based maintenance programs to reduce cost while improving uptime.
The advanced method: predictive maintenance
Predictive maintenance takes condition-based maintenance a step further. Once data is coming from equipment in real-time (or near real-time depending on each company’s needs), advanced analytics are used to identify asset reliability risks that could impact business operations.
By applying machine learning and analytics to operational data generated by critical assets to gain a better understanding of asset performance, companies can act on these insights as part of a continuous improvement process. In addition, data beyond machines can be used for predictions, such as weather data, information from other systems beyond traditional enterprise asset management systems, and any other data sources that may be valuable.
Companies with advanced processes and high-value equipment are rapidly adopting predictive maintenance solutions. But right now, these solutions aren’t for everyone – they require firms to have condition-based processes in place and are data intensive.
The future: prescriptive maintenance
Prescriptive maintenance is the future. It uses advanced analytics to make predictions about maintenance, but the difference is that prescriptive systems not only make recommendations but also act on recommendations.
Prescriptive maintenance requires that various asset management and maintenance systems are well integrated. For example, a predictive maintenance solution might recommend that a piece of equipment get overhauled based on analysis of vibration and temperature readings, but a prescriptive system would kick off a work order to field technicians based on this information and oversee the entire maintenance workflows.
Systems like this must be ‘cognitive’, or have the ability to think. This technology is at the intersection of big data, analytics, machine learning, and artificial intelligence. Companies such as IBM, with cognitive systems such as Watson and comprehensive enterprise asset management systems such as Maximo, are pioneering in this space.
By evolving from time based, to condition based, to predictive and prescriptive maintenance, companies are evolving their maintenance systems from being simply efficient to becoming truly strategic. Beyond maintenance, cognitive systems can integrate maintenance and operations data with other data sources, such as quality, warranty and engineering data, to become critical to how entire companies operate.