Fuzzy OEE offers manufacturers a way to examine performance variability when production data and operating conditions are not perfectly clear-cut. Overall Equipment Effectiveness (OEE) is widely used to assess how effectively equipment is producing, but its results can fluctuate with breakdowns, product changes, data collection practices and the period selected for measurement. Before pursuing improvement, teams need to understand whether these movements reflect real operational change, uncertainty in the inputs, or both.
A fuzzy logic approach does not replace conventional OEE. It extends analysis by representing uncertain inputs and losses in terms that accommodate degrees of membership rather than treating every value as precise. Used alongside Total Productive Maintenance (TPM), this approach can support more considered prioritization of production losses and corrective actions.
Why OEE fluctuates in production
OEE combines three dimensions of equipment performance: availability, performance and quality. Availability reflects whether equipment is ready to run; performance compares actual output with the expected rate; and quality accounts for acceptable output relative to total production. The familiar calculation is OEE = availability × performance × quality.
The measure helps make losses visible, but the inputs arise in a complex production environment. A change in any of the three components can move the overall result. The six major losses provide a practical lens: breakdowns and setup or adjustment time affect availability; idling, minor stoppages and reduced speed affect performance; and defects, rework and reduced yield during startup affect quality.
Fluctuations can also arise from uncertainty surrounding the process and its measurement. Relevant sources include:
- Equipment and process variation: Failures, variable lead times and changes in operating conditions can alter output from one period to another.
- Data collection: Manual records may be incomplete or inconsistent, while recording events can add a burden to shop-floor work.
- Loss classification: Teams may disagree about which OEE category best describes a stoppage or other loss.
- Measurement period: Daily values may show substantial variation, while longer reporting periods can average out day-to-day changes.
- Production context: Product mix, material availability, capacity and workforce conditions can all influence performance.
These factors show why a change in reported OEE should be interpreted in context. A single figure can conceal uncertainty about what happened, how it was recorded, or which losses contributed to the result.
Using fuzzy OEE to represent uncertainty
Conventional, or “crisp,” logic assigns an item to a category as either included or excluded. Fuzzy logic allows degrees of membership between zero and one. In production analysis, this can represent situations where a value or classification is not known with complete precision—for example, when the duration of a stoppage is approximate or its cause is not unambiguous.
Fuzzy OEE applies this idea to OEE calculations. Rather than assuming every input is an exact point value, the approach can represent inputs and equipment-dependent losses as fuzzy numbers. The resulting analysis acknowledges uncertainty in the data instead of hiding it behind a single precise-looking score.
In the FOEE method described in the source study, the calculation process broadly involves decomposing fuzzy inputs, converting their alpha-cut boundaries into arrays, evaluating the relevant expression, and recomposing the output. These steps provide a mathematical route for carrying uncertainty through the OEE calculation. The practical value is not simply a different score: fuzzy analysis can help relate variation in overall performance to the losses contributing to it.
Connect FOEE with TPM and shop-floor problem solving
TPM emphasizes shared responsibility for equipment effectiveness across production, maintenance and other relevant functions. Its practices include operator-led basic care, planned maintenance, training, focused improvement, quality maintenance, equipment development, office TPM, and attention to safety, health and the environment. This broad participation is important because OEE losses often involve interactions among people, technology and organizational practices.
A practical FOEE implementation can complement TPM through a structured improvement cycle:
- Map the process and clarify measurement rules. Define the equipment boundary, planned production time, data sources and rules for classifying losses. Consistent definitions make comparisons more meaningful.
- Identify uncertainty and loss sources. Review production records with operators, maintenance personnel and other stakeholders. Consider equipment behavior as well as the human and organizational conditions that influence data and operations.
- Prioritize consequential losses. Use the analysis to distinguish losses with a substantial influence on performance from issues that may be less relevant to the improvement objective.
- Develop and assess corrective actions. Multi-skilled teams can propose responses and consider their expected effect, including whether they address the underlying cause rather than only the reported number.
- Review results over time. Evaluate the process using a consistent measurement approach and revisit the assumptions, categories and data quality as operating conditions change.
The FOEE framework described in the study also considers the influence of individual losses and the assessment and selection of corrective actions. Experts may express expected effects using linguistic terms such as “high,” “medium” or “low,” where appropriate to the method. This can help structure judgment when precise measurements are not available, but it does not remove the need for sound data, agreed definitions and informed review.
Practical considerations before adoption
Fuzzy OEE should be treated as a complement to, not a substitute for, operational understanding. The method is most useful when teams can explain what is uncertain and why. If the underlying collection process is unreliable, adding fuzzy calculations alone will not resolve the root problem. Likewise, results depend on how the organization defines production time, categorizes losses and chooses the period for analysis.
Implementation should therefore begin with a clear purpose: identify which OEE fluctuations the team needs to understand and what decisions the analysis should support. A cross-functional review can then connect the numerical results with observations from the production line, maintenance history and operator knowledge. The approach outlined in the source study proposed FOEE as a framework for a food-processing case company, but did not complete the mathematical calculations or evaluate an improvement initiative over time. Its framework is consequently a guide for application, not evidence of a measured improvement outcome.
Conclusion
OEE remains a useful measure of availability, performance and quality, yet its fluctuations can be difficult to interpret when production conditions, data and loss classifications are uncertain. Fuzzy OEE provides a way to represent that uncertainty in the analysis and to examine how different losses influence overall performance.
For industrial teams, the practical implication is to pair the method with disciplined measurement and TPM-based collaboration. Establish clear definitions, involve the people who understand the process, prioritize meaningful losses and assess corrective actions systematically. This creates a more informed basis for stabilizing performance before pursuing further OEE improvement.