Predictive maintenance

Increase equipment reliability with predictive maintenance

The following demo is an example of how Rockwell Automation have built an integrated portal platform to transform their business.

With many of their assets in remote locations, it was previously very difficult and time-consuming to monitor this equipment, let alone understand its performance. By connecting them to Microsoft Azure, Rockwell Automation can now see real-time asset health and performance information, monitor service windows and carry out preemptive maintenance on parts and equipment before an incident occurs. The previously untapped wealth of data allows Rockwell Automation to analyse specific parts to determine service life or fine-tune optimal performance, transforming the way they, and their customers, do business.

These cloud services open up a variety of new business possibilities. For example, the equipment involved in mining, moving, refining and selling petroleum is expensive and rugged, and comes from hundreds of manufacturers. Enhanced by the Internet of Things (IoT), Rockwell Automation is extending its systems that monitor these valuable capital assets and use that data for predictive and even preventive maintenance. The Azure IoT solutions have the potential to transform the petroleum supply chain and produce bottom-line results in global productivity that could ultimately pay off at the pump.


KPI summary

Alerts and warnings

Asset details

Alert resolution

Step 1 of 3

Real-time remote monitoring

Operators can see the real-time location and health status of the entire infrastructure on the dashboard.

Step 2 of 3

View asset location and status

Understanding asset health is critical as any lost-time incident can be extremely costly to production volumes and supply contracts.

Step 3 of 3

Keep tabs on remote locations

Where a site might have been visited once every six months for routine maintenance, its status can now be tracked in real time.

Step 1 of 2

Track business metrics in real time

Critical site and production data rolls up to summarise important business KPIs, tracking them against targets and thresholds throughout the day or production period.

Get real-time results from assets that traditionally, may not be monitored for days, weeks or months

Step 2 of 2

Integrate with existing systems

Real-time sensor data can be combined with information from other external sources, or even other enterprise systems such as CRM, or ERP services.

Step 1 of 3

Real-time alerts and warnings

Alerts are escalated in real time for attention by the operator. Breakdowns can be responded to quickly through the custom portal or machinery shut-down through commands sent from the dashboard.

Step 2 of 3

Predict failures before they happen

Importantly, some alerts are predicted. Maintenance can be carried out in advance of a failure when data points to a situation, or trend, that the prediction model recognises as problematic.

Step 3 of 3

Resolve issues before they escalate

The operator can select the highest priority critical error that isn’t already being resolved or monitored.

Step 1 of 3

Analyse real-time data feeds

The dashboard ingests data in real time. At this level, the actual production performance and health status of an individual asset can be monitored.

Step 2 of 3

Empower decision-makers to take action

This data allows decision-makers to plan for scheduled (or unscheduled) work, organise maintenance windows or predict production outputs from assets that might be in remote locations.

Step 3 of 3

Conduct maintenance before an asset fails

In this example, a fan has a critical predicted warning. It will fail within days and cause the asset to be shut down. It is also well within the standard service life of this part. The operator can select the specific part to take action.

Step 1 of 5

Take action and resolve

The dashboard alert details provide the operator with specific information about the part and the recognised problem. This includes serial number, part number and even inventory and location of replacement items.

Step 2 of 5

Analyse business impact

The predicted failure notes that the air filter fan will fail before the unit is scheduled to have routine maintenance. This will result in a lost-time shut down of the asset.

Step 3 of 5

Analyse data in real time

Real-time data is ingested from devices in the field and displayed in the portal. The operator can monitor the real-time actual data feed to verify that the alerts and information provided are all correct. The threshold for the alert is also displayed so that the user can easily see the tracking in relation to normal operation.

Step 4 of 5

Create service tickets

The operator can create a ticket to have maintenance staff replace the part and keep the asset operational. They also have information and data to carry out business analytics and make any changes to operations based on the outcome.

Step 5 of 5

Create service tickets

The operator can create a ticket to have maintenance staff replace the part and keep the asset operational. They also have information and data to carry out business analytics and make any changes to operations based on the outcome.