It’s common knowledge in the manufacturing and processing industries that the costs of unplanned downtime are high and to be avoided as much as possible. That’s why there’s been so much focus on the value of asset management and predictive maintenance software to track end-to-end production performance and capture every event to highlight chokepoints in production.
The insights provided by such software applications are often touted as being among the quickest ways to achieve a good return on investment from early digital transformation initiatives.
According to Molly Culwell, applications engineer at dataParc, a supplier of data historian and visualization software, “When monitoring downtime, the goal is to gather meaningful metrics that will reveal why a process fails and how to best manage the problem in the future. At the very least, consider tracking which step failed, downtime duration, cause of the failure, the product or output affected, the crew on duty and other data points such as waste generated or safety issues.”
Armed with the insights provided by such software, you can establish a baseline of production performance using data from the software to create specific troubleshooting guides.
“Troubleshooting guides created using data captured from monitoring and maintenance reports can help get production back online quickly,” notes Culwell. “And using process dashboards to display workflows and analyses can highlight the root cause to make repairs faster.”
Determine which downtime cause is best to target and reduce. For example, a one-time occurrence due to a failed part is less critical than a recurring problem that stops productivity every month.
A key component of such troubleshooting guides involves use of the “five whys analysis" to determine the true source of a failure. “The five whys is a lean methodology designed to identify the root cause of a problem by asking the question “why” five times,” Culwell says. An example of the five whys can be seen in its application of a belt failure:
- Why did the belt fail? Because it was worn out.
- Why was it worn out? Because it wasn’t properly lubricated?
- Why wasn’t it lubricated? Because it wasn’t on the maintenance schedule?
- Why wasn’t it on the maintenance schedule? Because it was overlooked since no one knows how to maintain this machine.
- Why doesn’t anyone know how to maintain the machine? Because it is outdated and service training isn’t available.
In this example, asking the five whys shows the root cause of the failure to be aging, difficult-to-service equipment. Culwell notes that the long-term solution may be to replace this equipment. This is where applying the six-sigma tool known as DMAIC (define, measure, analyze, improve, control) can help in making that decision and avoiding equipment downtime overall.
According to Culwell, the application of DMAIC in terms of equipment maintenance applies as follows:
- Define. Create a list of potential reasons a process would be considered down, such as lack of pressure or a valve leak. Include both planned and unplanned causes. Use this data to inform you’re the maintenance and manufacturing software you use.
- Measure. Use the software to capture the who, what, when, where and why of an event. Use this data for the analysis phase and be sure to include comments from users.
- Analyze. Once the event is recorded and categorized, it can be analyzed. Determine which downtime cause is best to target and reduce. For example, a one-time occurrence due to a failed part is less critical than a recurring problem that stops productivity every month.
- Improve. Consider how to prevent and reduce downtime, such as by increasing the number of inspections.
- Control. Continue to monitor, measure and analyze to ensure that problems don’t reoccur. Adjust workflows and procedures as needed to maintain control.
