Automated production event data collection

Each piece of equipment can have several production counter tags assigned to it. Each production counter tag drive the capture of production counts for a specific disposition reason.

• The counter tag value will be an increasing value. The production count that is recorded will be the difference between the current value of the tag and the chronological previous value.

  • Each counter tag change might not result in a corresponding transaction in the production event data store. Counter changes might be accumulated and result in a single production event.

• The tag is mapped to a disposition reason in the efficiency model configuration. That disposition reason is used when recording the production count.

• The tag value should be a double or an integer. It can have an optional rollover value defined in the tags metadata (visible in the AVEVA Insight tag dictionary).

  • If the tag does not have a rollover, it will be a Reset counter (that is, it wraps from the last known value to 0 as soon as a lesser value is received).

  • If the tag does have a rollover, it will be a Rollover counter (that is, it wraps from the rollover value to 0 as soon as a lesser value is received).

    Rollover tag values should be within 10% of the rollover value before wrapping or be greater than 10% after wrapping. Otherwise, production counts can be missed.

    For more information about Historian counters and rollovers, see the Historian Server Scenarios Guide.

• When possible, tag values should be current and sent in chronological order. However, support for store-and-forward scenarios does exist.

  • Current data will typically be added as new production events to the current run-time context (that is, current production run, current OEE data property values).

    This behavior is based on the timestamp in the counter tag value, time, and quality (VTQ) indicator, so it is possible that it will be added to an older run-time context.

  • Old data (that is, store-and-forward data) will be attempted to be added as production events or to update production events in all the affected run-time contexts for the period for which the data was missing.

• Production counts are retrieved every 30 seconds if a counter tag update is received.

  If older data is detected (that is, when using store-and-forward), the data might take longer to process.

• VTQs with bad quality are ignored.