A performance threshold is a predefined numerical limit or boundary applied to a performance indicator that distinguishes acceptable from unacceptable operation. In industrial and manufacturing environments, it commonly refers to the value of a key performance indicator (KPI) or metric at which attention, investigation, or corrective action is expected.
How performance thresholds are used
In regulated and complex manufacturing operations, performance thresholds are typically defined for KPIs such as OEE, availability, yield, scrap rate, non-productive time (NPT), on-time delivery, or cycle time. These thresholds are often implemented in MES, SCADA, historian, or reporting systems to support:
- Monitoring and alerts: Triggering alarms, notifications, or visual flags when a metric exceeds or falls below the defined limit.
- Escalation and workflow: Initiating investigations, NCRs, CAPAs, or deviation workflows when performance crosses the threshold.
- Trend and risk analysis: Distinguishing normal variability from performance levels that may indicate quality, capacity, or compliance risk.
- Governance: Providing explicit, documented criteria for what is considered in-control vs. out-of-control performance.
Setting performance thresholds
Standards such as ISO 22400 describe terminology and calculation methods for manufacturing KPIs, but they generally do not define specific target values or performance thresholds. Organizations set their own thresholds based on factors such as:
- Product mix, technology, and equipment capability
- Historical performance and process variability
- Quality and regulatory expectations
- Risk tolerance and business objectives
- Existing MES/ERP configurations and validation practices
Thresholds may be single-sided (only a minimum or only a maximum) or two-sided (upper and lower limits), and can be static or adjusted over time as processes improve or change.
What performance thresholds are not
- They are not the same as the raw KPI definition or formula; they are limits applied to those KPIs.
- They are not guarantees of compliance or product quality; they are monitoring and decision boundaries.
- They are not always the same as contractual service levels or regulatory limits, although those may influence threshold settings.
Common confusion
- Performance threshold vs. target: A target is the desired performance level (for example, 85% OEE), while a threshold is the boundary that triggers attention or action (for example, investigation when OEE is below 80%). A plant may deliberately set thresholds different from targets.
- Performance threshold vs. specification limit: Specification limits apply to product or process characteristics (for example, part dimensions). Performance thresholds usually apply to operational metrics and KPIs. Crossing a performance threshold may lead to additional checks but does not automatically mean a product is out of specification.
Relation to ISO 22400 and KPI frameworks
Within KPI frameworks such as ISO 22400, performance thresholds sit on top of defined metrics like OEE, availability, and utilization. The standard helps ensure consistent calculation and interpretation of the KPIs themselves, while each organization defines its own performance thresholds and associated workflows in MES, ERP, and analytics systems.