How to reduce downtime in manufacturing with analytics

août 15, 2024

Mis à jour le: août 16, 2024

4 min de lecture

Analytics are a game changer when it comes to reducing downtime in manufacturing. They can reduce the length of unplanned outages, increase the lifespan of machines, and offer cost savings. Recent developments in the Internet of Things (IoT), big data, and machine learning are now coming together to enhance the abilities of these systems. This integration has led to great strides in tracking the deterioration of production equipment.

Modern solutions have made it possible for manufacturers to collect data from legacy SCADA systems and thousands of IoT sensors. The data provided here applies to hundreds of machines that spread across the factory floor. Analytics can then process this data in real-time, and compare it with historical data and virtual models. There are several different strategies for using these analytics to minimise downtime.

So, how can you use analytics to reduce downtime in manufacturing?

1) Condition monitoring analytics

Analytics can determine the present condition of a piece of machinery by keeping tabs on real-time data collected from its sensors. With this data and some measurement thresholds, condition monitoring analytics can find non-normal operating conditions.

Any emerging issues will trigger alerts and alarms for machine operators and engineers. After this, the equipment is carefully taken offline to ensure that proper maintenance can be carried out. This helps to prevent any unexpected downtime.

2) Advanced troubleshooting analytics

When a broken piece of equipment has already triggered a period of unscheduled downtime, analytics can reduce the length of time that the equipment is out of service. With the help of this software, you can remotely inspect the most recent machine data and try to identify the problem's source. Assisting engineers in identifying and fixing the likely source of the problem helps to reduce downtime.

3) Predictive maintenance analytics

With a large amount of historical data available, it is easy to see how analytics could optimise maintenance cycles. This data can predict failures that follow similar patterns and determine the best time to complete maintenance activities.

An operator or engineer can address the issue promptly. This prevents any unplanned downtime from occurring. When the real-time data closely matches established failure mechanisms, this allows for the possibility of this happening.

4) Commercial implementations of downtime analytics

The use of analytics to reduce downtime in manufacturing is still a form of new technology. Large engineering companies are now starting to introduce downtime analytics into their portfolios. Each company has its own unique approach to this.

Examples of companies using downtime analytics

1. Schneider Electric has an EcoCare service membership which uses condition-based maintenance to reduce planned downtime by up to 40%. Also, it extends asset life cycles through the use of dynamic maintenance scheduling [1].
2. Siemens has a predictive maintenance platform that creates machine and maintenance models using artificial intelligence. It says that it can reduce unplanned downtime by up to 50% and reduce maintenance costs by up to 40% [2].
3. FANUC has developed an IoT-based downtime analytics solution that monitors the condition of automotive manufacturing robots. It uses the results of this analysis to predict maintenance requirements and reduce downtime. From an installed base of 35,000 robots, it has avoided 1,700 instances of unplanned downtime [3].
4. Deloitte has a predictive maintenance framework based on IoT sensors that uses downtime analytics to predict asset failures. It has a chance to extend the life of assets and cut downtime by between 5 and 15% [4].

There are many different approaches that may be used to reduce downtime in manufacturing by using analytics. Combining several types of analytics can make them more effective. It is possible to reduce downtime by using analytics to monitor the condition of equipment in real-time. Then, troubleshoot problems when downtime occurs, and undertake preventative maintenance simultaneously.

[1] “New EcoCare Services Membership from Schneider Electric Helps to Reduce Modular Data Centers’ Planned Downtime by Up to 40%”, 4 April 2023, Schneider Electric

[2] “Senseye Predictive Maintenance”, https://www.siemens.com/global/en/products/services/digital-enterprise-services/analytics-artificial-intelligence-services/predictive-services/senseye-predictive-maintenance.html, Siemens

[3] “ZDT (Zero Down Time)”, https://www.fanuc.eu/uk/en/iot-solutions/zdt, FANUC

[4] “Predictive maintenance Deloitte's approach”, Deloitte