What Is Overall Equipment Effectiveness (OEE) and How AI Improves It

Overall Equipment Effectiveness (OEE) is a key performance indicator that measures how efficiently manufacturing equipment and processes run. It is calculated as OEE = Availability × Performance × Quality, and AI-powered vision systems can automate much of the manual data entry the calculation requires.

By visually monitoring machines and production lines, these systems can detect machine states (running, idle, down), count parts (both good and defective), and even calculate cycle times directly from video footage. This significantly reduces the need for human input, increasing data accuracy and freeing up operators for more value-added tasks.

Investing in AI-powered OEE initiatives drives significant value and return on investment (ROI), leading to cost savings, waste reduction, and improved competitiveness in the market.

Maximizing Manufacturing Efficiency with OEE and AI

OEE is one of the most important metrics for optimizing operational efficiency in manufacturing. It has been around as a concept for decades but historically relied on manual inputs and estimates. AI now makes it a practical metric to track accurately.

What is Overall Equipment Effectiveness (OEE)?

What is OEE? OEE is a key performance indicator (KPI) used to assess the efficiency of manufacturing equipment and processes. It provides insights into how well equipment is utilized, how productive it is during scheduled production time, and how effectively it performs when running. OEE is calculated based on: Availability, Performance, and Quality.

The formula for OEE is: OEE = Availability × Performance × Quality.

OEE Definitions

• Availability: This measures the percentage of time that equipment is available for production, accounting for factors like downtime for maintenance and unplanned stoppages.
• Performance: Performance indicates the speed at which equipment operates compared to its maximum potential speed, considering factors like equipment running at reduced speed and minor stoppages.
• Quality: Quality measures the percentage of good-quality products manufactured compared to the total number of products produced, accounting for defects, rework, and scrap.

For example, if your equipment is available 24x7x365, running at maximum capacity, and never produces a defect, you would calculate this as:

100% Availability * 100% Performance * 100% Quality = 100% OEE

Even the best equipment will sometimes need maintenance, so this is an unrealistic example, but it helps you understand the concept. 

Calculate your OEE

What is a good OEE score?

Today, there seems to be a consensus that an OEE score of 85% is “world-class”. This number may increase as more manufacturers move towards “lights out” or “closed-loop” systems. 

To put an 85% OEE in perspective, if your equipment is available 95% of the time, running at 90% of capacity, and producing high-quality results 99% of the time, you would calculate your OEE as follows: 

95% Availability * 90% Performance * 99% Quality = 84.6% OEE

While you can round that up to 85% OEE, it technically isn’t world-class. Sources vary on other OEE score benchmarks. Some claim the average OEE is 60%, while others believe it ranges between 65-85%. In reality, it varies depending on your specific sub-industry. What matters most is that you accurately measure your own OEE and create actionable strategies to improve it. 

How does your OEE compare?

How does AI improve OEE measurement?

The concept of OEE is not new, but measuring it accurately has always been challenging. Historically, OEE calculations relied on workers manually measuring, reporting, and entering data. Accuracy and consistency depend on who is measuring, reporting, and entering data at any given time.

AI eliminates the variables and ensures that all data considered is accurate and trustworthy. This consistency allows the AI to create more accurate measures of efficiency and more useful recommendations on improving OEE over time. Vision AI, along with today’s soft sensor technologies, can gather precise information on uptime, performance, and defect rates. Vision AI systems use machine learning algorithms to inspect products in real time, identifying defects more accurately than humans and reducing the number of false positives. By integrating Vision AI into production lines, manufacturers can ensure consistent product quality, minimize defects, and boost OEE.

What are the challenges of measuring OEE manually?

Multiple, essential manual data entry points are needed to calculate OEE on a single process. Manual tracking, data entry, and calculations become exponentially more challenging to complete for complex manufacturing with multiple processes to measure, and the risks of inaccurate data increase. 

Here are some of the variables that must be tracked for each process:

Time-Based Data

It’s vital to track downtime reasons separately for more in-depth analysis (e.g., breakdown vs. changeover).

• Planned Production Time: The total scheduled time the process is expected to operate during a shift or specific period.
• Run Time: The actual time the process was actively producing parts. This excludes planned breaks, maintenance, etc.
• Downtime: Any period where the process was not running due to:
• Breakdowns: Equipment failures or malfunctions
• Changeovers: Time spent switching from one product to another
• Setups: Time for initial adjustments and calibration
• Minor Stoppages: Brief interruptions in production
• Reduced Speed: Instances of the process running slower than ideal

Uptime vs. downtime of an overflow line in plastics molding facility

Count-Based Data

• Total Count: The total number of parts produced during the period analyzed.
• Good Count: The number of parts meeting all quality specifications.
• Defect Count: The number of parts with defects or requiring rework.

Vision AI detects and counts rods & strand while assessing quality for OEE calculation inputs

Other Important Data

Ideal Cycle Time: The theoretical fastest time the process can produce a single good part without defects.

Even with manual OEE data collection, using a spreadsheet or a simple OEE calculation tool will be much more efficient and less prone to error than doing calculations by hand.

What is the value of improving OEE with AI?

Investing in OEE improvement initiatives powered by AI can yield significant value and return on investment (ROI) for manufacturers. Because of the number of variables, it is hard to create definitive ROI in a blog post, it is best to speak with an expert to get a tailored estimate. Some of those variables include:

1. Increased Equipment Availability: Predictive maintenance powered by AI reduces unplanned downtime, ensuring optimal utilization of manufacturing equipment.
2. Improved Production Efficiency: AI-driven performance analysis identifies and eliminates bottlenecks, optimizing production processes and enhancing equipment performance, leading to improved efficiency.
3. Enhanced Product Quality and Compliance: Vision AI-based quality inspection ensures consistent product quality, reducing defects, rework, and scrap. This not only improves customer satisfaction but also helps manufacturers comply with quality standards and regulations.
4. Cost Savings and Waste Reduction: Minimizing downtime, improving efficiency, and reducing defects result in substantial cost savings and waste reduction, directly impacting the bottom line.
5. Competitive Advantage: Manufacturers that effectively utilize AI to optimize OEE gain a competitive edge by being more agile, efficient, and responsive to market demands.

Example ROI:

A line that produces 60 units per minute for 18 hours a day, 7 days per week. Margins per unit are $1.00.  Starting OEE is 40%.  After measuring and understanding areas for improvement, OEE is increased to 45%, resulting in over $1.3M in additional value.

OEE is a critical metric for measuring and improving operational efficiency in the manufacturing industry. While it was introduced as a concept over 30 years ago, it is poised to take on even more prominence thanks to advancements in AI and precision tracking. Now, manufacturers can increase their trust in OEE and unlock new opportunities for optimization. 

By investing in AI-powered OEE initiatives, manufacturers can improve equipment availability, production efficiency, and product quality, and gain a competitive advantage in their industry.

Calculate the incremental value of OEE improvements

Frequently asked questions

How do you calculate OEE?

Multiply Availability by Performance by Quality. For example, equipment that is available 95% of the time, runs at 90% of capacity, and produces good parts 99% of the time has an OEE of 84.6%.

What is a good OEE score?

An OEE of 85% is generally considered world-class. Sources vary on averages, with some citing 60% and others 65-85%, and the right benchmark depends on your sub-industry.

How does Vision AI help measure OEE?

By visually monitoring machines and lines, AI-powered vision systems detect machine states (running, idle, down), count good and defective parts, and calculate cycle times from video, which reduces manual data entry and improves accuracy.

What data do you need to calculate OEE?

Time-based data such as planned production time, run time, and downtime by reason; count-based data such as total, good, and defect counts; and the ideal cycle time for the process.