- Artificial Intelligence
- Work Order Management
Work order management: How to go from corrective to predictive maintenance
The Internet of Things (IoT) and big data allow for real-time analysis and equipment status monitoring.
The result? Predictive maintenance lets service managers anticipate faults, extend equipment life, and optimize work order management.
Now that the cost of connected objects has dramatically decreased, multiple sensors to measure temperature, humidity, vibration or pressure are being placed on equipment in ever greater numbers.
By working together, IoT and big data are enabling the automated detection of equipment failure before it happens, making this after-sales service dream a reality.
From 1980s remote monitoring to today’s big data
Manufacturers first introduced sensors on machinery in the late 1980s. Later, in the early 2000s, 24/7 remote monitoring and surveillance systems were deployed. Since then, the computing power and scalability of cloud-based networks have been deployed en masse. That’s important because connected objects produce huge volumes of data.
Consider this fact: a single sensor recording a measurement each and every second is capable of generating a staggering 31-million readings each year. So, for continuous real-time monitoring and feedback a high-powered big data infrastructure is a must.
Working in real time
Big data also includes the capability for real-time analysis of time-stamped sensor data. This round-the-clock monitoring allows service managers to detect equipment performance changes, including availability and latency.
And, because machine status data can now be gathered at any time and in real time, there’s no more waiting for a technician site visit.
Welcome to Industry 4.0
The concept of fourth generation manufacturing, also known as Industry 4.0, brings together the best of IoT, big data, and cloud computing. This coming together can extend a device’s life in two ways: by preventing breakdowns and by improving device function/operation.
Moreover, by developing a deep understanding of the equipment and how it functions, the operator can configure it optimally. This, in turn, can reduce energy consumption.
The limits of corrective and preventive maintenance
When it comes to the limits of corrective and preventive maintenance, a couple of points need to be made:
Predictive maintenance is different from corrective and preventive maintenance.
With corrective maintenance, equipment repair occurs once a fault is detected. Typically, the equipment is stopped, or at least slowed down, making service response time and return to normal operations critical.
For its part, preventive maintenance means the equipment is maintained at regular intervals. This approach, based on the estimated life cycle of the equipment components or parts, relies on scheduled precautionary replacement of spare parts.
From statistical to real data analysis approach
Unlike the statistical approach, predictive maintenance is based on the analysis of real data. This way the service manager is sure that the spare part shows no sign of wear and that the replacement can be avoided or postponed.
This paradigm shift — from preventive to predictive maintenance — optimizes inventory management, as well as valuable technician time. And, equally important, the predictive maintenance approach allows equipment operators to minimize machine downtime and production slowdowns.
Many benefits but a significant investment
Predictive maintenance can achieve the pinnacle of after-sales service: just-in-time (JIT) service. This is when field technicians are called upon to intervene only when needed ― based on hard evidence.
What’s more, predictive maintenance improves the resolution rate at first contact, also known as the First Time Fix Rate. That’s because the deployed field technician knows the equipment and its maintenance history.
To summarize, predictive maintenance has many advantages, including:
Limited risk of serious failure
Lengthened equipment service life
Better work order planning
Reduced equipment downtime
Optimized spares inventory management
On the other hand, predictive maintenance requires a greater up-front investment than corrective and preventive maintenance. That’s because operators must set up, then manage the connected objects, as well as the data processing and data warehousing infrastructure.
Artificial intelligence (AI) enables the leap forward
According to a four-country study by Vanson Bourne, commissioned by GE Digital, 75% of IT decision makers and management believe that, by 2020, machine health status will be better managed than the status of human health.
Also, Gartner predicts that by 2020, 10% of field service visits will be triggered and dispatched by an AI-enabled device. For this reason, alongside IoT and big data, AI is the third pillar of the coming after-sales service revolution.
In fact, it’s already possible to schedule work orders on the fly with algorithmic models that take into account technician availability, traffic conditions, and ordering spares. Soon, AI platforms will advise the service manager whether or not to initiate particular work orders.
To get there, what’s needed is greater platform autonomy. It won’t be long before intelligent machinery will be able to self-diagnose and, based on its health status, trigger actions, such as downloading a fix, reconfiguring for reduced output, requesting assistance, or ordering spare parts. It will be as if the machines are speaking their own maintenance language.
Our similar articles.
- Field Service Management
Technology increases productivityJuly 3, 2018
- Internet of Things
- Field Service Management
- Work Order Management
- Work orders
5 ways Artificial Intelligence is making Field Service Management smarter (Part 2)April 12, 2018
Why maintenance is so important in the air conditioning industrySeptember 3, 2019