In 2014, Lenny Liebmannin used the term DataOps, a concept that combines data science, DevOps, and agile principles, which is used today in industrial settings for smart machine maintenance and reliable and efficient delivery of products. It helps manufacturing companies turn industrial data into tangible value while handling their processes, automation, and workflows. The principles of Industrial DataOps–automation, focus on quality, real-time analytics, transparency, and collaboration–combined with the optimized approach to machine monitoring, help in the seamless integration of data from various sources (sensors, log files, production data, telemetry) to empower companies to improve their manufacturing processes.
This necessitates the need for a smart machine monitoring platform–built in alignment with the principles of DataOps and powered by artificial intelligence–to help ingest, monitor and analyze machine data to enable proactive maintenance, increase uptime, enhance safety with predictive maintenance, improve compliance with industry standards and data protocols , and assist in informed decision-making.
The intelligent use of machine data enables today's organizations to build more efficient and resilient operations. This machine data from connected operations, although available in abundance, is often difficult for manufacturing companies to generate value from. This is where a smart machine monitoring solution/system–such as Rawcubes’ iDataOps–helps operationalize data in enterprises.
What exactly is OEE and how does it relate to DataOps in industrial settings–also referred to as Industry 4.0?
Overall Equipment Effectiveness (OEE) is a measure of machine availability, performance, and quality. It is a valuable metric that helps in evaluating the effectiveness of a machine maintenance strategy. Here’s what each variable means:
Availability measures the percentage of time that the machine runs to produce a certain output. For example, if the runtime of a machine is 8 hours, but the maintenance team shuts it down for 1 hour for maintenance, then its availability would be 7/8. That would be 87.5%.
Performance measures the speed at which the machine produces an output, let’s say widgets. For instance, if a machine is designed to produce 100 widgets per hour(capacity), but it can only output a maximum of 90 widgets per hour, then its performance becomes 90%.
Quality measures the percentage of products that the machine outputs without defects and is ready for shipping. For example, if a machine produces 100 widgets, but 15 of them turn out to be defective, then its quality would be 85/100. That would be 85%.
We can calculate the ROI of machine monitoring systems by comparing the costs of the system to the benefits it provides. The benefits of machine monitoring systems can include:
Prevention of unplanned downtime
Reduced maintenance costs
The costs of machine monitoring systems can include:
The cost of the system itself
The cost of installation and configuration
Though the ROI of machine monitoring systems can vary depending on the specific application, their benefits always outweigh the costs.
Let’s take an example to understand this.
Let's say a machine has a planned production time of 1 hour. The machine is available for 50 minutes and runs at 90% of its ideal speed. However, 10% of the parts produced are defective.
Based on this info, the variables would take the following values:
Availability: Available time / Total production time = 50 minutes / 60 minutes
The OEE for this machine would be:
OEE = Availability * Performance * Quality
= 50 minutes / 60 minutes * 0.9 * 0.9
This means that the machine is only 67.5% effective. There are opportunities to improve the OEE by reducing downtime, increasing speed, or improving quality. We’ve already established that the effective use of a machine monitoring platform can help reduce downtime.
Now, let's say that the machine monitoring platform can increase the machine’s availability to 55 minutes, i.e. 10%. This is possible through the real-time monitoring and predictive maintenance features that a machine monitoring platform offers. Through real-time monitoring, the maintenance team can obtain real-time alerts on an asset’s condition. This makes it easier for the team to resolve the problem swiftly. The machine monitoring platform also collects and analyzes data from machines to identify potential problems. This helps the team schedule maintenance activities. Coming back to our example, let us assume the use of the solution would increase the availability to 55 minutes / 60 minutes = 0.91.
The new OEE would be:
OEE = 0.91 * 0.9 * 0.9
This means that the OEE has increased to 73%. The following table summarizes the improvement in OEE:
|Original OEE||New OEE||Improvement|
If we extend this example, and plug in a couple numbers, this is what we have. Let’s assume the machine produces 60 units per hour running at full capacity. The cost of producing each unit is $100. In the first case, the machine runs for 50 minutes, producing 50 units. This generates about $5000 in revenue per hour. After using a machine monitoring platform, the machine runs for 55 minutes, effectively producing 55 units. This increases the revenue from $5000 to $5500. This is an additional $500 profit per hour. In a year, the machine would run throughout the working days (assuming there are 260 working days), considering it can run efficiently for 10 hours a day. This brings the profit amount to $13,00,000. (Please note that this is a fictitious example)
This shows that the improvement in OEE can lead to significant financial benefits, which explains the growing need for machine monitoring platforms in manufacturing industries.
Here are some benefits of using a machine monitoring platform:
Early detection of problems: It can help identify problems early before they cause unplanned downtime. This can save businesses money on repairs and cost production.
Improved decision-making: It can provide businesses with insights into machine performance. This information can be used to make better decisions about maintenance, scheduling, and production planning.
Increased productivity: It can help businesses to identify and eliminate inefficient processes. This can lead to increased productivity and profits.
Improved equipment reliability: It enables condition-based maintenance, predicts potential failures based on historical data, and minimizes unexpected breakdowns, leading to enhanced equipment reliability.
Improves resource allocation: It empowers floor managers to plan the utilization of resources, such as machines, labor, and materials optimally.
Traditional techniques, such as manual inspections and scheduled maintenance, can only identify problems after they have occurred. In contrast, machine monitoring software can help improve ROI in several ways. By monitoring machine performance in real time, manufacturers can identify potential issues before they occur, allowing for proactive maintenance and reducing downtime. This can lead to significant savings, as downtime can cost manufacturers hundreds or even thousands of dollars per hour. It also assists manufacturers in improving OEE by identifying and addressing inefficiencies in their production processes.
Take a proactive approach to machine maintenance with iDataOps' predictive maintenance from Rawcubes. Contact Rawcubes today to discover how iDataOps can give your organization the advantage it needs and transform it into an Industry 4.0. smart factory.