Manufacturers face the constant challenge of accurately measuring and tracking performance to remain competitive. Without precise metrics, it is difficult to identify bottlenecks, justify capital investments, or uncover the hidden factory - the untapped potential of existing equipment that could be unlocked without purchasing new assets.
The three core metrics of OEE, OOE, and TEEP solve this by providing different zoom levels of performance data, ranging from tactical shop-floor efficiency to high-level strategic capacity planning.
OEE measures tactical efficiency during scheduled production (Did we run well?). OOE measures operational effectiveness over an entire shift (Did we use our staffed time well?). TEEP measures strategic capacity over 24/7 calendar time (What is our total potential?).
Use OEE to improve daily floor operations and TEEP to decide if you need to buy new equipment or simply add a new shift. Computer vision fixes the Quality component of OEE at the source rather than at the exit.
1. Overall Equipment Effectiveness (OEE)
Overall Equipment Effectiveness is the most widely used performance indicator in the manufacturing industry. It measures how effectively a machine or process operates during its planned production time. The formula for calculating OEE is Availability × Performance × Quality.
Key Components
Availability: The ratio of run time to planned production time. It accounts for unplanned stops like breakdowns and lack of operators.
Performance: A measure of how fast the machine ran compared to its ideal cycle time. It highlights losses from micro-stops and slow cycles.
Quality: The ratio of good parts produced to the total number of parts.
Example
If a machine is scheduled for 8 hours but breaks down for 1 hour, its availability is 87.5%. If it runs at 90% of its rated speed and produces 2% scrap, the OEE would be approximately 77%.
Use Cases
OEE is best for day-to-day operational excellence. It helps supervisors track the impact of specific losses like changeovers and machine failures during active shifts.
2. Overall Operations Effectiveness (OOE)
Overall Operations Effectiveness takes a broader perspective than OEE by measuring effectiveness against the entire shift time or total operational time. The formula for calculating OOE is Performance × Quality × Availability (where Availability = Actual Production Time / Operating Time).
What it Includes
Unlike OEE, which ignores pre-arranged downtime, such as lunch breaks and scheduled maintenance, OOE counts these as part of the total time the machine could have been running.
Example
Consider a machine that runs for 6 hours during an 8-hour shift. OEE might be 75% (measured against the 6 scheduled hours), but OOE would be 56% (measured against the full 8-hour shift).
Use Cases
OOE is ideal for workforce and resource management. It helps planners understand how much capacity is lost to organizational choices like staffing levels or shift patterns.
3. Total Effective Equipment Performance (TEEP)
Total Effective Equipment Performance is the most comprehensive metric, measuring equipment performance against all available time (24 hours a day, 365 days a year). The formula for calculating TEEP is OEE × Loading (or Utilization Rate). Loading is the ratio of scheduled time to total calendar time.
Key Idea
TEEP is designed to reveal the maximum potential of a facility. If a plant has low TEEP but high OEE, it means the equipment is efficient but largely idle on nights or weekends.
Example
A machine with 75% OEE running a single 8-hour shift, 5 days a week, has a Loading factor of roughly 23%. Its TEEP would be only ~17%, showing that 83% of its annual capacity is untapped.
Use Cases
TEEP is a strategic tool for executive leadership. It is primarily used for capital investment decisions - determining whether to buy a new machine or simply add a second shift to an existing one.
How to Use Each Metric Correctly
Choosing the right metric depends on the question you are trying to answer.
Frequently Asked Questions
Q. Why is my OEE higher than my OOE?
OEE only counts the hours you planned to work. OOE counts the entire shift, so it includes more down time (like breaks), which naturally lowers the percentage.
Q. What is a world-class score?
For OEE, 85% is considered world-class.
For TEEP, world-class is typically 80% or higher, though this is usually only achieved in continuous 24/7 process industries like oil or paper mills.
Q. Can OEE exceed 100%?
The overall OEE cannot, but the Performance component can exceed 100% if a machine runs faster than its ideal design speed.
Q. Should I track TEEP for every machine?
No. TEEP is most valuable when applied to bottleneck machines that limit your total throughput. Applying it to machines with excess capacity often creates unnecessary noise.
How Computer Vision Increases the Quality Component of OEE
As you’ve seen, in the OEE framework, Quality is calculated as the ratio of good units produced to total units started, meaning every defective part that reaches end-of-line inspection (or worse, escapes to the customer) directly erodes your Quality score. Computer vision attacks this problem at the source rather than at the exit.
The Core Mechanism: Inline Inspection vs. Downstream Sampling
Traditional quality control relies on statistical sampling, such as inspectors pulling parts at intervals, checking against go/no-go gauges, or reviewing end-of-shift batches. The fundamental flaw is latency. A tooling drift that begins at 10:00 AM may not surface in a sample pull until 11:30, producing 90 minutes of scrap in the interim.
Computer vision systems mounted inline (at the press, the weld station, the assembly cell, or the conveyor) inspect every unit, every cycle, in milliseconds. The moment a defect signature appears (a burr, a void, a misaligned component, a surface scratch, a solder bridge), the system flags or ejects that part and, critically, can trigger an alert or even a line stop before the next cycle begins.
Computer vision catches defect categories in real-time
- Dimensional variance - Using structured light, laser profilometry, or stereo vision to detect out-of-tolerance features that gauging would catch but far more slowly.
- Surface defects - Scratches, pits, porosity, discoloration, and contamination detected through high-resolution 2D imaging with controlled lighting.
- Presence/absence verification - Confirming that fasteners, labels, O-rings, connectors, and other components are correctly seated before the part advances.
- Assembly sequence errors - Verifying correct part orientation, insertion depth, and mating of subcomponents.
- Weld and solder quality - Detecting undercut, porosity, incomplete fusion, bridging, and cold joints that X-ray or manual inspection would catch only in audit.
The OEE Quality Math
Suppose a line runs 1,000 units per shift and currently operates at 94% Quality (60 defective/rework units). Implementing inline computer vision that catches defects at Station 3 rather than at end-of-line - and corrects the upstream cause immediately - might reduce that defect volume to 20 units. Quality moves from 94.0% to 98.0%. On a line running two shifts, that delta represents 80 recovered good units per day with no increase in Availability or Performance.
The compounding effect is significant. Parts caught early are often reworkable or allow die/tooling correction before an entire batch is compromised. Parts caught at shipping or by the customer carry a multiplied cost, including warranty claims, field service and reputational damage, that never appears in an OEE calculation but destroys the economics behind it.
Need help improving OEE Quality?
Roboflow empowers developers and enterprises to build their own fast, accurate, and scalable computer vision applications. We help you automate quality inspections, see real-time insights, identify bottlenecks, and elevate manufacturing yield and equipment effectiveness, all while reducing costs. Book a demo.
Cite this Post
Use the following entry to cite this post in your research:
Contributing Writer. (May 4, 2026). Understanding OEE vs OOE vs TEEP. Roboflow Blog: https://blog.roboflow.com/oee-vs-ooe-vs-teep/