ISO 22400 distinguishes raw data, indicators, and KPIs primarily by their level of processing, context, and decision relevance. It does not dictate your plant’s exact metric list, but it gives a reference model to structure how you define and govern metrics in manufacturing operations.
Raw data
In ISO 22400, raw data is the lowest level of information:
- Directly captured values, typically from sensors, controllers, PLCs, MES, ERP, or manual entry.
- Examples: cycle start/stop timestamps, part counts, temperature readings, operator IDs, material lot IDs, alarm codes.
- Minimal or no transformation applied beyond basic validation and formatting.
- Often time-stamped and associated with a resource, order, material, or operation, but not yet aggregated into performance measures.
Key point: raw data by itself is usually not what management or quality leadership uses for decisions, but it is the traceable evidence base from which indicators and KPIs are derived. In regulated environments, retaining this level with appropriate genealogy and audit trails is critical for reconstruction and investigations.
Indicators
ISO 22400 uses “indicator” as a broader category of processed measures that sit between raw data and KPIs:
- Derived or transformed from raw data by calculation, classification, or aggregation.
- Express a state or performance aspect of equipment, a work center, a line, or an order.
- Frequently include time-based aggregation (per shift, per order, per hour) or state aggregation (e.g., time in setup vs production vs downtime).
- Can be leading or lagging, and may be very local (e.g., per machine) or more global (e.g., per plant).
Examples aligned with ISO 22400 concepts:
- Availability ratio for a line or machine.
- Performance ratio based on ideal vs actual cycle time.
- Quality ratio based on conforming vs nonconforming output.
- Mean time to repair (MTTR) and mean time between failures (MTBF).
- Planned vs unplanned downtime proportions.
In practice, an indicator in ISO 22400 is any defined metric that results from processing raw data and that can be reused in analysis, comparison, and decision-making. Many indicators can exist even if only a subset are elevated to KPIs.
KPIs (Key Performance Indicators)
KPIs in ISO 22400 are a subset of indicators that are explicitly designated as “key” because of their management and business relevance:
- They are indicators that have been selected and agreed as central to monitoring manufacturing performance against business objectives.
- They typically roll up multiple indicators or express a particularly important dimension (e.g., OEE or on-time delivery rate).
- They are intended to support regular review by operations, quality, and management, not just local troubleshooting.
- They usually have defined targets, thresholds, and escalation rules.
Examples in the ISO 22400 family of concepts:
- Overall Equipment Effectiveness (OEE), usually calculated from availability, performance, and quality indicators.
- On-time in-full (OTIF) for manufacturing orders or deliveries.
- Scrap rate or rework rate at plant or value-stream level.
- Adherence to production schedule for critical product families.
The standard does not force you to adopt a fixed set of KPIs. It defines reference KPIs and their relationships to underlying indicators and data, but each organization still has to choose which indicators are actually “key” for its context and risk profile.
How the three levels relate in practice
A typical chain in an ISO 22400-style model looks like this:
- Raw data: Machine reports 100 cycles completed, 5 failed tests, timestamps for each cycle, and downtime events with reason codes.
- Indicators:
- Good parts = 95, defect count = 5.
- Availability ratio = operating time / planned time.
- Performance ratio = ideal cycle time / actual cycle time.
- Quality ratio = good parts / total parts.
- KPI:
- OEE = availability × performance × quality.
- This KPI is used in weekly operations reviews and tracked against a target.
From a governance standpoint, ISO 22400 encourages clear, documented relationships between these layers so that when a KPI moves, you can trace back to the supporting indicators and ultimately to the raw data and equipment states.
Constraints and dependencies in real plants
How cleanly you can apply the ISO 22400 distinctions depends heavily on your environment:
- System coexistence: Legacy MES, historians, SCADA, and ERP may each calculate their own indicators. Aligning definitions and avoiding duplicate or conflicting KPIs requires explicit modeling and cross-system data governance.
- Data quality: If timestamps, counts, or state logs are incomplete or inconsistent, the same KPI definition can yield different values by system. ISO 22400 does not solve this; it only provides a structure to define and compare metrics.
- Integration and validation: In regulated or aerospace-grade contexts, any change to KPI logic, data sources, or aggregation periods can trigger revalidation, documentation updates, and change control. This is a significant reason full replacement of metric-calculation logic across all systems is rarely done in a single step.
- Physical constraints and long equipment lifecycles: Older assets may not provide all the raw data ISO 22400 assumes. You may need proxies, manual data collection, or partial implementations, and should document these limitations explicitly.
In summary, ISO 22400 gives a conceptual hierarchy: raw data at the base, indicators as processed measures, and KPIs as selected, business-critical indicators. Each organization still has to map its existing brownfield data and systems into this structure, with careful attention to traceability, validation, and consistency.
