One of the most important aspects to remember during data collection is that the operating state of the system must be the same every single time. This is quite difficult to do, however. So when the analyst sees "high" sensor data, they may incorrectly assume that it is a system fault. It may have just spiked, however, because of increased loads.
FREMONT, CA: Predictive maintenance (PdM) is no longer a good thing to have in today's competitive industrial world; it has become a necessity. Traditional PdM methods have a number of limitations. However, advancements in cloud, wireless, and AI technology have disrupted the way PdM has been done in recent decades. Companies are using these technologies to provide an end-to-end, easy-to-deploy PdM solution at an incredibly affordable price point. This pattern is known as the Industrial Internet of Things (IIoT). Predictive maintenance is currently the most commonly used case for IoT in all industries.
Let us look the main reasons why a walk around program does not constitute a truly predictive maintenance solution.
Stay ahead of the industry with exclusive feature stories on the top companies, expert insights and the latest news delivered straight to your inbox. Subscribe today.
Failures between data collection rounds: When walking around programmes, including well run ones, technicians only collect data once every 30 or 90 days. The problem is that any errors that arise after the last data collection cycle will remain undetected before the next data collection cycle. As a result, computers remain vulnerable to failure, resulting in unplanned downtime.
Check Out: Top Predictive Maintenance Solution Companies
Different operating conditions: One of the most important aspects to remember during data collection is that the operating state of the system must be the same every single time. This is quite difficult to do, however. So when the analyst sees "high" sensor data, they may incorrectly assume that it is a system fault. It may have just spiked, however, because of increased loads.
Inconsistent data collection: During a walk around the programme, technicians could not position the sensors precisely and consistently. This leads to incorrect data and incorrect estimation of the health of the machine.
Inaccessible machinery: Machines are sometimes not easily accessible, either because there are safety issues or because they are behind a cage. So technicians end up not collecting data on machines that cause them to malfunction.
Manual Analysis: All data obtained must be analysed manually. It's not real time, and it's hard to scale up manual research when you're working with hundreds and thousands of devices.
More in News