Let’s say you take your car in for an oil change. The oil you select is rated for traveling a distance of 5,000 miles. In all likelihood the technician places a stickeron the car’s front window – a simple reminder of when the oil will be due for another change. In roughly 4 months this routine will be repeated. That’s an everyday example of Preventive Maintenance – a simple time-based approach to maintaining costly assets.
Now imagine that you purchase a new car – one that has built-in system diagnostics. From the time you first drive away from the dealership it actively tracks the accumulation of miles. What’s more it evaluates the outside temperature, driving conditions, load, and amount of idling. The car’s algorithm assesses all of these factors and proactively alerts you to the need for an oil change. It might occur at the 5,000 mile market, or it might not. That’s an example of Predictive Maintenance – an advanced approach to protecting assets.
Predictive Maintenance stands apart from more traditional and preventive approaches in that it doesn’t rely on time as the sole, deciding factor. As the name suggests predictions in an industrial setting are made when maintenance of plant equipment are assessed as necessary. While time is often a key factor predictions are based on other important and measurable equipment attributes. This approach to maintaining equipment is increasingly the norm across the process industries.
If controller performance monitoring technologies are also predictive, then how are they related to the Predictive Maintenance solutions available in the market? Here are several aspects that link the two:
Control loop monitoring tools provide alerting mechanisms similar to their predictive maintenance counterparts. More specifically, most loop monitoring tools employ a combination of real-time alerts and scheduled reports. Users subscribe to the tool’s alerting and reporting functions that align with their specific job function or area of responsibility. Together these capabilities assure that important information is delivered on time and that appropriate issue ownership is established.
As with predictive maintenance technologies, key performance indices (KPIs) enable controller performance monitoring tools to accurately measure the health of plant assets – in this case assets are individual PID loops and the associated mechanical equipment. Representative control loop KPIs include those that evaluate mechanical issues like Stiction as well as PID controller challenges such as oscillation. These KPIs proactively measure control loop attributes that are key to production.
Predictive maintenance solutions assign thresholds that trigger alarms if and when those thresholds are surpassed. Similarly loop performance monitoring tools set specific thresholds to the various KPIs so that appropriate staff are alerted when performance slips into inefficient and/or harmful territory. They serve notice that either an investigation or an adjustment is warranted in order to reestablish safe, efficient control.
The average production facility has 100s if not 1000s of control loops so it should be expected that controller performance monitoring technologies rank each
identified issue from high to low. That’s no different than most predictive maintenance solutions. Just as some assets have a higher value than others in a plant, so too the performance of some PID loops outweigh others in terms of their impact on production. By prioritizing issues loop monitoring products assure that attention is appropriately paid to the loops that matter most.
Clearly there are numerous similarities between controller performance monitoring technologies and Predictive Maintenance solutions. Both provide manufacturers with valuable advance warning. While similar in nature they assess the health of different parts of a production process. As with the oil in your car there are plenty of things at a plant that can go wrong and that can bring production to a halt.