5Ps of Control System Reliability and Performance–Part 3: Predictive Maintenance

The 5Ps of Control System Reliability--Predictive MaintenanceIn Part 2 of this series, we briefly mentioned the concept of Predictive Maintenance. We continue here with a deeper examination of its benefits, components, and considerations for its implementation.

Monitoring the health of your industrial control system is vital to ensure its reliability and performance. Continuous, or at least frequent, periodic monitoring leads to early detection of potential failure points and enables the automation teams to address them proactively. This proactive stance results in fewer failures, reduced repair time, and an overall increased lifetime of your system assets. Furthermore, maintenance activities are optimized, allowing for the spacing of preventive maintenance routines, ultimately leading to better resource allocation and cost savings.

Deploying a predictive maintenance solution involves both physical and intangible components. At a high level, here are some things to consider:

  1. Define Objectives and Procedures: This step involves identifying the goals and key results for implementing the monitoring system in the first place. Determine which aspects and components of the control system will be monitored, define the procedure for incident response, and assign ownership of alerts or faults that will be detected.
  2. Select or Develop Tools: Once objectives have been defined, identify the tools capable of collecting relevant data. Since the control system will have various components (computers, controllers, networking equipment), select applications and sensors that can connect to and monitor these components.
  3. Configure Data Collection: Software agents, sensors, and devices must be configured to capture relevant parameters automatically. More importantly, these monitoring components should not disrupt or interfere with the system’s normal operation.
  4. Setup Threshold and Alerts: Define thresholds and rules to trigger alerts/notifications, along with the mechanism for sending the alerts (email, SMS, HMI banners). A reliable alerting mechanism would ensure that timely notifications are sent to responsible stakeholders.
  5. Develop Reports and Visualization: Develop dashboards that provide real-time insights into the system’s health. These act as a viewing window to the enormous amount of data and parameters being collected or monitored. The data presentation must also be intuitive enough for actionable information to be easily derived.
  6. Review and Refine: Periodically review the effectiveness of the monitoring system to assess whether the investment is delivering the defined goals or just generating additional work and nuisance alerts. Remember that a fundamental objective of setting up the system is to reduce the amount of work (and not add to it!) This calls for periodic assessment of the parameters being tracked and the thresholds that have been set. The technologies being used must also be reviewed as the complexity of the process or system evolves.

Given the considerations above, it is apparent that implementing a continuous monitoring system for predictive maintenance requires multi-faceted expertise: in distributed control systems, network infrastructure, software development, and data analytics. Depending on their desired results, system owners can choose whether just to surface the diagnostics information already built into the various system components, build their own full-fledged monitoring system with alerting and data visualization capabilities, or explore readily available solutions that were developed with these objectives in mind.

Emerson has made the choice simple for its customers by introducing the DeltaV Predictive Maintenance solution. This continuous monitoring solution helps reduce dependence on manual tasks and increases awareness of system diagnostics, accelerating response time to events.

Deployment Option for DeltaV Predictive Maintenance

Deployment Option for DeltaV Predictive Maintenance (Click to enlarge)

A possible deployment of the system is depicted above. This system automatically performs hundreds of health checks on DeltaV system components, including controllers, workstations, servers, safety systems, switches, firewalls, UPS units, and even virtualization infrastructures, as well as non-DeltaV workstations. The collected data is aggregated and visualized in a highly intuitive dashboard that displays the monitored assets along with the alerts that have been generated.

DeltaV Predictive Maintenance provides sufficient features and flexibility to meet all the considerations we outlined earlier. Installation is straightforward, data collection is automatic, and health checks and thresholds are fully configurable. Users also have the option to disable or snooze alerts, minimizing the occurrence of nuisance alerts during startup or shutdown system states.

What sets the solution apart, however, is its ability to identify probable root causes and provide recommendations for addressing abnormal conditions detected by the system. An example is shown below.

Probable Cause and Recommended Action for an Alert

Probable Cause and Recommended Action for an Alert (Click to enlarge)

With this unique capability, enabled by Emerson’s deep knowledge and expertise in the DeltaV control system, the time between issue identification and resolution is greatly reduced. This directly translates to cost savings in the form of downtime avoidance and optimized resource allocation. Moreover, this knowledge base also equips automation teams with the essential know-how that would have otherwise had to come from long hours of training in various disciplines such as DCS, networking, programming, and data analytics and visualization.

Transitioning from Reactive and Preventive Maintenance to Predictive Maintenance is a strategic move toward enhancing the availability and performance of control systems. The proposed approach above can help you optimize your maintenance activities and resource allocation. Emerson’s DeltaV Predictive Maintenance solution, in particular, simplifies this process and empowers the automation team with actionable insights that could lead to significant cost savings and performance improvement.

Author’s Note:

While some solutions linked above are mainly for Emerson DeltaV control system users, the concepts and principles should generally apply to any other Industrial Control System. If you want to learn more:

The post 5Ps of Control System Reliability and Performance–Part 3: Predictive Maintenance appeared first on the Emerson Automation Experts blog.