IMPLEMENTING SHOP FLOOR ANALYTICS: MACHINE MONITORING TELLS ALL
Many manufacturers fail to get the maximum amount of productivity out of their machine tools. In fact, according to research by Machine Metrics, the average machine utilization rate in 2018 was less than 29 percent – but most manufacturers are oblivious to the fact that there’s a problem with their shop’s efficiency. Luckily, the solution is simple – as Dr. Mikel Harry of the Six Sigma Academy stated, “We don’t know what we don’t know. We can’t act on what we don’t know. We won’t know until we search.” For manufacturing, that means getting a handle on productivity by identifying inefficiencies in the manufacturing process, and the best way to accomplish this is with machine monitoring and modern shop floor analytics.
In the past, the only way to analyze a shop’s efficiency was by gathering the data by hand, a cumbersome process that could take days or weeks for a basic evaluation of machine utilization at a medium-sized shop. While manual data gathering might make sense for a shop with fewer than five machines, automated machine monitoring speeds this process up at an extraordinary rate for larger shops. Simply implementing real-time feedback, for example, can improve productivity by up to 20 percent.
Implementing a machine monitoring solution has been transformed into a plug-and-play process with today’s machine tools. However, even legacy equipment rarely requires more than adding a few sensors and configuring a simple input/output adapter. A typical IoT gateway device will allow linking machines and ERP/MES systems together using WiFi, ethernet or a cellular network, which then passes the data along to a cloud computing platform that stores, aggregates and analyzes it.
Of course, even the sophisticated array of sensors available in modern machine tools are unable to capture all of the data shops need to gain a full understanding of their productivity. Instead of relying on handwritten reports, however, machine monitoring solutions now offer tools like touchscreen human-machine interfaces that give operators and maintenance specialists the ability to add human context to data. By identifying what caused a period of downtime or why a part was rejected, operators can make it much easier to separate isolated problems from systemic issues.
Machine performance data are used for a variety of applications after they are collected and aggregated. The easiest way to use this data is through the use of key performance indicator (KPI) dashboards that track machine utilization in real time. In addition to showing operators how they are performing as they work, the data are also used for large-screen displays that show how entire machining cells or production floors are performing – a powerful tool for identifying bottlenecks and potential issues as they occur.
Of course, with the best machine monitoring solutions, it’s possible to go further than speeding up reaction times – artificial intelligence and machine learning algorithms are increasingly making it possible for manufacturers to predict maintenance needs ahead of time. Simple rules-driven predictive alerts are fairly easy to set up and utilize condition monitoring and preventive maintenance schedules to reduce downtime. But with rapid advancements in AI development, true predictive maintenance is becoming a reality.
Learn more about how easily you can integrate machine monitoring into your operations here.
Inline Content - Survey
Current code - 5fce8e61489f3034e74adc64