{"id":93402,"date":"2026-05-16T17:05:42","date_gmt":"2026-05-16T17:05:42","guid":{"rendered":"https:\/\/teeptrak.com\/availability-performance-quality\/"},"modified":"2026-05-16T17:05:43","modified_gmt":"2026-05-16T17:05:43","slug":"availability-performance-quality","status":"publish","type":"post","link":"https:\/\/teeptrak.com\/en\/availability-performance-quality\/","title":{"rendered":"Availability, Performance, Quality: how are the three OEE components defined?"},"content":{"rendered":"

[et_pb_section fb_built=”1″ _builder_version=”4.27″][et_pb_row _builder_version=”4.27″][et_pb_column type=”4_4″ _builder_version=”4.27″][et_pb_text _builder_version=”4.27″]<\/p>\n

Availability, Performance, Quality: how are the three OEE components defined?<\/h1>\n

Last verified: 16 May 2026.<\/strong> Overall Equipment Effectiveness (OEE) is the product of three components: Availability, Performance, and Quality. Each is independently defined in ISO 22400-2:2014 (Automation systems and integration \u2014 Key performance indicators (KPIs) for manufacturing operations management \u2014 Part 2: Definitions and descriptions) \u00a76.4, with the underlying time and quantity elements specified in \u00a75. The combined framework originates in Seiichi Nakajima’s 1988 Introduction to TPM<\/em> (Productivity Press, ISBN 0-915299-23-2), with the German VDI 3423:2011-08 standard providing the parallel European definition.<\/p>\n

\n

“Together, they produce availability, performance, and quality with consistent definitions that match the standard.”<\/em> \u2014 OPC Foundation Machine Tools Companion Specification, 2021, \u00a7C.1.4, citing ISO 22400-2:2014.<\/p>\n<\/blockquote>\n

The three components are conceptually independent: each isolates a different family of loss. A line can have high Availability and low Performance (it runs when scheduled but slowly), or high Performance and low Quality (it runs at speed but produces defects). The multiplicative composition into OEE makes the metric sensitive to whichever component is weakest, which is the design intent of the original Nakajima framework.<\/p>\n

Availability<\/h2>\n

Availability = Actual Production Time (APT) \u00f7 Planned Busy Time (PBT)<\/strong><\/p>\n

Availability measures the fraction of scheduled production time that the equipment was actually running. Per ISO 22400-2:2014 \u00a75.3:<\/p>\n

    \n
  • Planned Busy Time (PBT)<\/strong>: the time during which the equipment is intended to be operating, excluding scheduled non-production time (breaks, planned maintenance, planned changeovers in some definitions).<\/li>\n
  • Actual Production Time (APT)<\/strong>: PBT minus all unplanned downtime events (equipment failure, setup time over plan, adjustment time).<\/li>\n<\/ul>\n

    The Six Big Losses framework (Nakajima 1988, Chapter 3) maps Losses 1 (Equipment Failure) and 2 (Setup and Adjustment) directly to the Availability component. Improvements to Availability require attacking either MTBF (failure frequency) or MTTR (repair duration), per the steady-state identity Availability = MTBF \u00f7 (MTBF + MTTR) given in MIL-HDBK-338B \u00a75.8.<\/p>\n

    Performance<\/h2>\n

    Performance = (Planned Run Time per Item \u00d7 Produced Quantity) \u00f7 Actual Production Time<\/strong><\/p>\n

    Performance compares actual output rate to the design rate for the part being made. The denominator is APT (the same quantity used for Availability). Per ISO 22400-2:2014 \u00a75.2:<\/p>\n

      \n
    • Planned Run Time per Item (PRI)<\/strong>: the design cycle time per part for the specific product being run. Sourced from the MES or product specification.<\/li>\n
    • Produced Quantity (PQ)<\/strong>: the total number of parts produced during APT, including both good and defective.<\/li>\n<\/ul>\n

      Six Big Losses 3 (Idling and Minor Stoppages) and 4 (Reduced Speed) map to Performance. Note the distinction from Availability: a stop under the threshold (typically 5 minutes) is not counted as downtime; instead, it appears as reduced Performance because the cycle did not produce a part during that interval. This is the operational reason micro-stoppage detection requires sub-second sensor sampling.<\/p>\n

      Quality<\/h2>\n

      Quality = Good Quantity (GQ) \u00f7 Produced Quantity (PQ)<\/strong><\/p>\n

      Quality measures the first-pass yield. Per ISO 22400-2:2014 \u00a75.4, Good Quantity is the quantity that meets requirements in the first time of an operation process<\/em>. Rework that brings parts back into specification does not count toward GQ; this is the most frequently violated definition in practice.<\/p>\n

      Six Big Losses 5 (Defects and Rework) and 6 (Startup Losses) map to Quality. The framework treats a defective part the same whether it is scrapped or successfully reworked \u2014 both reduce the Quality ratio. Vendors reporting Quality above 99% should be challenged on whether rework is being excluded from GQ as the standard requires.<\/p>\n

      Worked example demonstrating component independence<\/h2>\n

      Consider three CNC machining lines with the same OEE = 60%, decomposed differently:<\/p>\n\n\n\n\n\n\n\n
      Line<\/th>\nAvailability<\/th>\nPerformance<\/th>\nQuality<\/th>\nOEE<\/th>\nDominant Loss<\/th>\n<\/tr>\n<\/thead>\n
      A<\/td>\n67%<\/td>\n95%<\/td>\n94%<\/td>\n60%<\/td>\nBreakdowns or long setup<\/td>\n<\/tr>\n
      B<\/td>\n92%<\/td>\n69%<\/td>\n94%<\/td>\n60%<\/td>\nMicro-stoppages or speed<\/td>\n<\/tr>\n
      C<\/td>\n88%<\/td>\n91%<\/td>\n75%<\/td>\n60%<\/td>\nDefects or rework<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n

      The three lines have identical OEE but require entirely different interventions. Line A needs maintenance work to address MTBF and MTTR. Line B needs sub-second sensor measurement and root-cause investigation of micro-stoppages \u2014 this is the pattern we observed across the Hutchinson deployment (TeepTrak customer, 40 plants in 12 countries, OEE moved from 42% to 75%). Line C needs quality engineering attention to first-pass yield. Reporting OEE alone provides none of this signal.<\/p>\n

      The boundary between Performance and Quality is sometimes unclear<\/h2>\n

      A subtle measurement issue arises when reduced speed produces defects, or when in-process inspection rejects parts mid-cycle. Different implementations of ISO 22400 handle this differently:<\/p>\n

        \n
      1. Strict ISO 22400-2:2014 normative version: defective parts count as Produced Quantity, reducing Quality. They also count toward the throughput rate, reducing measured Performance loss.<\/li>\n
      2. Nakajima 1988 strict interpretation: defective parts do not count as Produced Quantity at all; they appear only in the Quality component as the difference between PQ and GQ.<\/li>\n<\/ol>\n

        The Iannone and Nenni 2020 peer-reviewed analysis (DOI: 10.1016\/j.cie.2020.106660) documents this as one of the inconsistencies within ISO 22400 that practitioners must resolve at the measurement layer. TeepTrak’s default is the Nakajima 1988 interpretation; the ISO 22400 strict interpretation is available as a configuration option for audit contexts.<\/p>\n

        Time element hierarchy under ISO 22400<\/h2>\n

        The four-level time hierarchy from Kang et al. (NIST Pub. 919754) and ISO 22400-2:2014 \u00a75 frames the three components:<\/p>\n

          \n
        • Calendar Time \u2192 Planned Operation Time \u2192 Planned Busy Time \u2192 Actual Production Time \u2192 Net Operating Time \u2192 Value-Added Time<\/li>\n
        • Availability is the ratio APT \u00f7 PBT (one level)<\/li>\n
        • Performance is the ratio Net Operating Time \u00f7 APT (next level)<\/li>\n
        • Quality is the ratio Value-Added Time \u00f7 Net Operating Time (lowest level)<\/li>\n<\/ul>\n

          The three components together compress the time hierarchy from PBT down to Value-Added Time; OEE is the overall compression ratio.<\/p>\n

          Frequently asked questions<\/h2>\n

          How are Availability, Performance, and Quality calculated?<\/h3>\n

          Availability = APT \u00f7 PBT. Performance = (PRI \u00d7 PQ) \u00f7 APT. Quality = GQ \u00f7 PQ. All per ISO 22400-2:2014 \u00a76.4.<\/p>\n

          Why is the OEE the product, not the sum?<\/h3>\n

          The components measure different families of loss applied sequentially: the line must first be available (Availability), then run at speed (Performance), then produce good output (Quality). Multiplication captures the compounding nature of these losses.<\/p>\n

          Can one component exceed 100%?<\/h3>\n

          Under correctly applied ISO 22400-2 definitions, no. Performance values above 100% typically indicate the PRI is set shorter than physically achievable.<\/p>\n

          How do the Six Big Losses map to A\/P\/Q?<\/h3>\n

          Losses 1-2 affect Availability, Losses 3-4 affect Performance, Losses 5-6 affect Quality.<\/p>\n

          What is the Nakajima 1988 world-class benchmark for each component?<\/h3>\n

          Availability \u2265 90%, Performance \u2265 95%, Quality \u2265 99%, giving an OEE benchmark of approximately 85%.<\/p>\n

          Does rework count toward Good Quantity?<\/h3>\n

          No. ISO 22400-2:2014 \u00a75.4 defines GQ as parts meeting requirements in the first time of an operation process. Reworked parts do not count.<\/p>\n

          Should planned maintenance reduce Availability?<\/h3>\n

          No. Planned maintenance is excluded from PBT (the denominator), so it does not reduce Availability. Only unplanned downtime within the planned production schedule reduces Availability.<\/p>\n

          How do micro-stoppages affect Performance?<\/h3>\n

          Micro-stoppages (typically under 5 minutes) reduce Produced Quantity during APT without changing APT itself, so they appear as reduced Performance rather than reduced Availability.<\/p>\n

          What is the difference between ISO 22400 and VDI 3423?<\/h3>\n

          Both define equipment availability and related KPIs. VDI 3423:2011-08 is the German engineering standard with European industrial-equipment heritage; ISO 22400-2:2014 is the international standard with broader scope. The numerical results are typically equivalent for discrete manufacturing.<\/p>\n

          How does TeepTrak measure A\/P\/Q?<\/h3>\n

          From the sensor state stream and production counters. Availability from running-state transitions, Performance from cycle-rate comparison against PRI, Quality from downstream inspection or reject counter integration. Each component is independently reported alongside the composite OEE.<\/p>\n

          References<\/h2>\n
            \n
          1. ISO 22400-2:2014. Automation systems and integration \u2014 Key performance indicators (KPIs) for manufacturing operations management \u2014 Part 2: Definitions and descriptions<\/em>. International Organization for Standardization, Geneva.<\/li>\n
          2. Nakajima, S. (1988). Introduction to TPM: Total Productive Maintenance<\/em>. Productivity Press. ISBN 0-915299-23-2.<\/li>\n
          3. Kang, N., Zhao, C., Li, J., and Horst, J.A. (2016). A Hierarchical structure of key performance indicators for operation management and continuous improvement in production systems. NIST Publication 919754.<\/li>\n
          4. OPC Foundation (2021). Machine Tools – Companion Specification<\/em>, OPC UA version 1.02, \u00a7C.1.4.<\/li>\n
          5. VDI 3423:2011-08. Availability of machines and production lines \u2014 Terms, definitions, time categories and calculation<\/em>. Verein Deutscher Ingenieure, D\u00fcsseldorf.<\/li>\n
          6. Iannone, R. and Nenni, M.E. (2020). Overall Equipment Effectiveness: consistency of ISO standard with literature. Computers and Industrial Engineering<\/em>, vol. 145, 106660. DOI: 10.1016\/j.cie.2020.106660.<\/li>\n<\/ol>\n

            Author: Bastien Affeltranger, CTO, TeepTrak. Reviewed by Fran\u00e7ois Coulloudon, CEO. Cross-references: OEE<\/a>, Six Big Losses<\/a>, MTBF<\/a>. Last verified 16 May 2026 against ISO 22400-2:2014.<\/em><\/p>\n