Key Performance Indicators

Overview

KPI Index

• Efficiency
  • Target Efficiency
  • Lifetime Efficiency
  • 1 hour Efficiency
  • 24 hour Efficiency
  • 1 week Efficiency
• Cycle Time
  • Target Cycle time
  • Lifetime Cycle Average
  • 1 hour Cycle Average
  • 1 week Cycle Average
  • Last Cycle Interval
• OEE
  • Target OEE
  • 1 hour OEE
  • 1 week OEE
• Temperature
  • Baseline Temperature
  • Lifetime Max Temperature
  • 1 hour Max Temperatore
  • 24 hour Max Temperature
  • 1 week Max Temperature

Rules

1. Hourly KPIs are exactly 1 hour long and start at the top of every hour. Therefore there are 24 hourly KPI buckets in a day.
2. Time-based KPI analysis (e.g., 1-hour, 24-hour, 1-week) are performed on the current period. This means that in a 1 hour bucket, if you are 15 minutes into an hour, your 1-hour OEE will show the OEE for that 15 minute period going back to the top of the current hour.
3. If a tool had no cycles in a bucketed period, the value will be empty for that period. This means that the 1 hour efficiency column will be empty in a report if the tool has not run within the current hour.

KPI Evaluation Process

1. Every cycle we get has a timestamp.
2. Based on that timestamp, identify what specific kpi interval of type DAY , HOUR and SHIFT it falls into. Each "kpi interval" has a start and end date. For example, DAY kpis start when the plant day starts and end when the next day starts.
3. If there is no KPI entry for that interval, create a new one
4. Apply changes to the metrics in the KPI interval (e.g., increment the downtime ).
5. Save the updated KPI back into the database.

Example: Computing Hourly Statistics

1. Look in the saved KPI data for the asset, for the entry of type hour that "started" in the last hour. For example, if it's now 10:33, is there a KPI entry that started after 09:33.
2. The system finds the next possible hourly bucket, which is the 10:00 bucket.
3. Compute the statistics using the data in that horuly bucket.