Ten new releases have been launched over the past twelve months for the innius platform. These have successively added to and improved functionality such as; virtual sensors, shift visualisation, customizable dashboards, demo machines, configurable KPIs and a responsive UI. But despite all the fantastic progress, arguably this latest release, incorporating data forecasting and known as release 1.11, is the most significant to date.
What’s the big deal?
Until now innius has excelled at offering functionality which offers greater real-time insight into machine behaviour and gives stakeholders the means to respond quickly. This means that innius works really well at supporting machine operators who are using a Condition Based Maintenance (CBM) strategy. But release 1.11 includes functionality which can forecast sensor data values for up to twenty-four hours in advance and trigger warnings if this forecast meets predefined rules. This functionality is significant because it breaks away from Condition Based Maintenance and is the first small step into supporting Predictive Maintenance (PdM) strategies.
A weather analogy
A good analogy for data forecasting is weather forecasting. We often have to make our plans based on the weather. Is it likely to rain? Shall I cycle to work or go by car? Shall I take an umbrella? The first thing we usually do is look outside to see the current conditions. It’s nice and sunny. But that says little about the rest of the day. We then look at the weather forecast and find that this afternoon there will be heavy downpours. We adapt our plans accordingly and decide we’d better drive than cycle.
In a similar way, machine data forecasting allows industrial machinery managers to have a good expectation of how a machine will perform during the coming 24 hours, and adapt operational plans accordingly. And also like weather forecasts we have to consider that the longer the forecast the less accurate it is, so for now innius is offering forecasts of a maximum of 24 hours to ensure accuracy.
Unexpected breakdowns, misdiagnosis of problems and unnecessary repairs are the bane of industrial machine manufacturers and operators. A Condition Based Maintenance strategy can go quite far in addressing these issues, especially when using a smart platform such as innius. But Predictive Maintenance strategies, like weather forecasts, have the ability to go much further by enabling informed decisions and changes to behaviour in advance. That’s why Predictive Maintenance is the ultimate goal for machine manufacturers and operators; and why innius wants to support it.
Practical examples of data forecasting:
- innius forecasts an increase in bearing vibration, which indicates that a conveyor is about to break down. The operations manager is notified and can plan a convenient time to have this repaired.
- innius forecasts an increase in defects, which indicates a quality issue with supplied materials. The operations manager is notified and production of the batch can be stopped immediately, saving costs.
This latest release is just the first small, but significant, step for innius into supporting Predictive Maintenance, let’s wait and see what follows!
Find out more
To find out more about data forecasting refer to the following Knowledge Base articles:
How to setup analytics forecasting for a sensor
How to view analytics forecasting