Request a Quote

Thank you! Your request has been sent. We'll be in touch soon.
Oops! Something went wrong while submitting the form.

What is a Measure?

In Timebase, the "pieces of information" you want to track are modeled as Measures. They are the PIs (performance indicators) or KPIs (key performance indicators) against which Timebase collects and transforms data into information. The information stored by a Measure is numeric but allows for the addition of attributed data for context.

As data streams into Timebase, it is cleaned, contextualized, transformed and stored against a Measure. The name and position of the Measure in the Timebase Model provide its model context. Examples of Measures include:

  • Performance - volumes, yield, efficiency, quality
  • Asset Management - running hours, stroke counts, availability, utilization, downtime, OEE
  • Inventory - raw materials, intermediates, warehouse stock
  • Cost and usages - water, electricity, waste, rework
  • Personnel - safety, incident management, absenteeism, time management
  • Projections - projected production, projected revenue

What is an Event?

As part of the time context on which Timebase information is based, Event periods are derived from triggers in the underlying data. Timebase monitors for start and stop triggers that generate periods dynamically. For example, Timebase will monitor the necessary tags, or a combination of tags, to record when a machine stops and starts up again.

In Timebase, the definition of these triggers is modeled as an Event. Additional information, like the reason for the stop, can also be defined as part of that Event. As data streams into Timebase, trigger expressions are analyzed and event period information generated is stored against the Event. Like a Measure, the name and position of the Event in the Timebase Model provide its model context. Examples of Events include:

  • Production - batches, lots, work order runs, production runs
  • Schedules - planned production, shifts
  • Machine State - downtime, OEE, cleaning in place
  • Threshold monitoring - data stream input above or below a threshold

Sizing a Timebase System

Sometimes it isn't easy to know how many Measures and Events a Timebase System needs. A typical business objective (e.g. Improve Filler 1's efficiency by 5%) would require between 5 and 10 Measures for inputs and calculations, as well as 1 or 2 Events for machine state and downtime monitoring. Some business objectives may not need Event definitions at all.

In general, a Timebase System will have more Measures than Events by at least one order of magnitude. Please get in touch with us for advice on sizing your Timebase System. And remember, you can always add Measures (in bundles of 100) and Events (in bundles of 10) to your Timebase System as your information needs grow.

Need some advice?