Do you ever feel there’s never enough time in the day for what needs to be done versus what could be done? If you are a floor manager, you know exactly what needs to be accomplished during a shift to keep lines running and products moving out the door.
On the good days, this runs like a well-oiled machine. That is, until something, often unexpected, brings things to an untimely halt. That could be a line that remains down because of a part held up in the supply chain, a staffing shortage, or a worker injury.
With time and people stretched like never before, it is no longer viable to rely on a fingers crossed, reactive form of maintenance. When time is money, it becomes a dangerous choice to wait until a piece of equipment no longer works. Too often, this form of non-action holds the potential for unplanned downtime and an increase in collateral damage to surrounding components.
Amid trying to keep lines up and running and things on track, it can be easy to dismiss the long-term impact predictive maintenance can have on a facility. Predictive maintenance looks to short-circuit a problem before it happens. This could include removing components before they reach their full use as well as using data and monitoring to predict the lifetime of components and equipment. These actions can reduce the potential of a failure that could result in downtime and offer insight in how to reach the larger goals and desired ROI of the organization.
Ask the Questions
- Is it possible a failure could occur within X hours?
- How many hours could be left in the life of the asset?
- Is the asset performing/behaving normally?
- Which asset needs service first?
Apply the Data
Along with answering these questions, preventative maintenance efforts can be further enhanced by data collected from algorithms, sensors, operator interactions and feedback, and environmental inputs. With the right measures in place, a manufacturer can benefit from receiving data before, during and post-failure.
Through the application of these data points, manufacturers have a range of knowledge (cloud computing, data analytics, IoT and machine learning) to support their preventative maintenance efforts. By applying historical knowledge, statistical modeling, and simulations to predict known failure rates, manufacturers can create a long-term advantage for themselves and their customers.