KbA0016: Working Rule Execution and Insight/CTL Generation

Description

The purpose of this KbA is to discuss how often working rules run, and how soon Insights and calculated Trend Logs get generated.

Each Working Rule created under a Rule Template requires associated input Trend Logs/constants to be mapped to the input Logic Blocks before it can be executed.

When does my rule run? How soon can I expect Insights and Calculated Trend Logs to be generated?

In general, there are two types of Rules, Triggered and Scheduled. Triggered Rules use polled Trend Log inputs, whereas, scheduled Rules have COV Trend Log inputs.

Triggered Rules

Working Rules run when new samples are available for all the input Trend Logs. For instance, let’s assume we have a Working Rule that has two inputs – one is the Chiller kW TL and the other is an Outside Air Temperature (OAT) Trend Log. This rule will run next when new samples are available for both the inputs – Chiller kW Trend Log and OAT. The rule will not run again until new samples arrive in Kaizen for both the Trend Logs.

Trend Log data must be ‘current’ to trigger a rule. ‘Current’ is defined as data from the previous day or today. This means expired data will not trigger the Rule to run automatically, however, you can execute the Rule manually in this case.

Scheduled Rules

COVs will generate a new sample only when the state of the monitored object changes. Therefore if one (or more) of your inputs is a COV Trend Log, and if there are no new samples available, it is assumed that the state has not changed, and the rule will be processed using the last sample received. In other words, rules run on a schedule using the most recent data available.

Scheduled Rules will run 1- 2 times each day.

Manually running Working Rules

Even though the Working Rules are processed based on the availability of new samples for input Trend Logs, there would be use cases where you might want the Working Rules to be processed immediately. One example would be when you write a Working Rules to identify all the times a particular fault condition had occurred in a system for the last year and current year to date – you might want to view the information sooner than 4 hours since you need this info to be used elsewhere, in which case you can manually run the Working Rules.

Please refer to the How To Manage Working Rules page for more information about this. When a Working Rule is run manually, it does not wait for new samples to arrive: it is run based on the existing data for all the input Trend Logs. Additionally, the Working Rule goes right to the front of the analytic engine’s queue for quicker processing. When the Working Rule is processed successfully, the associated insights and Calculated Trend Logs are generated.

Receiving outputs (Insights and CTLs)

CopperTree generates Insights and Calculated Trend Logs using best-effort delivery. Once a rule is triggered, CopperTree’s infrastructure has been designed to generate the outputs (Insights and Calculated Trend Logs) and deliver them as quickly as possible. Typically, this is done within minutes (generally less than an hour) of new samples arriving in Kaizen for all the input Trend Logs or within seconds for manually executed Working Rules. However, we could run into cases where we’re unable to deliver within that window. This can be caused by a substantial influx of data, a server fault or error, or other unknown/unforeseeable issues. In such cases, CopperTree’s system support staff will respond to issues promptly and are dedicated to resolving them as quickly as possible.

If you have any questions or concerns, please contact CopperTree Customer Solutions at [email protected]