Automating condition monitoring and the benefits for manufacturers

By Dr Simon Kampa, CEO of Senseye

The introduction of machine-learning and Industry 4.0 technologies has increased the scalability and accessibility of condition monitoring, accelerating adoption across the manufacturing sector. The potential savings of both time and money through reducing downtime and making maintenance more efficient have made condition monitoring an attractive, (Read More)  feasible and almost obligatory aspect of industry.

The origin of data-driven condition monitoring

Data-driven condition monitoring is by no means a new concept in the manufacturing world. For 25 years or more it has been mandatory in the aerospace and defence industries, due to heavy regulation and the high safety standards expected of them. However, gathering and analysing sufficient data to drive meaningful results has been an expensive, time-consuming and laborious task. This has limited investment from other manufacturing sectors.

The Industry 4.0 movement, and the introduction of the cloud and machine-learning technologies is now changing the game. Smart sensors and machines that can read their own vital statistics have automated the collection of machine data, while software such as Senseye is able to identify patterns and anomalies in machine health by analysing this data.

By reading machine data outputs around aspects such as vibration, temperature and the amount of electrical current drawn, it is now possible to assess the health of machines meticulously using a computer, spot emerging problems up to six months before they might affect production, and put in place more precise, more appropriate maintenance schedules.

A key factor in performing this analysis is the use of machine learning algorithms that are sufficiently generic to be used on any instrumented machine from any manufacturer, enabling the assessment of machine condition to begin immediately. The accuracy of this analysis improves exponentially over time as we discover more from the data about each machine’s unique quirks and characteristics.

This automated, predictive approach to assessing machine condition and organising maintenance scheduling is proving transformative for manufacturers. Engineers can look at a simple dashboard each morning to see where their efforts would be best applied and when, enabling them to minimise downtime by ensuring machines are scheduled for maintenance weeks before a predicted failure.

Benefits of automated analysis

Unplanned downtime is one of the biggest costs for any manufacturing environment. In the automotive sector, for instance, every minute that critical machinery carries a cost amounting to tens of thousands of pounds.

Being able to predict if and when a machine will fail in the future allows engineers to make repairs weeks or even months before a predicted failure might affect production. Senseye customers have halved their levels of machine downtime and cut their maintenance costs by up to 40 percent using this automated approach.

While traditional approaches to condition monitoring required organisations to recruit data scientists, automation means manufacturers can achieve the benefits of condition monitoring without having to find these increasingly scarce and expensive skills. Those that already employ data scientists can now leave mundane day-to-day monitoring to the computer and allow the human experts to focus their attention on the most complex issues that require more creativity and lateral thinking.

While an automated approach to prognostics and condition monitoring represents a new way of working for many, the benefits more than justify the change. Given the financial and operational benefits of introducing predictive maintenance at scale are now so much greater than the cost of implementing and managing such programmes, it is hard to conceive any large scale manufacturer not using them in the decades to come.