AstraZeneca recently piloted a new data collection system on a packaging line at its Westborough, MA facility to collect data to identify ways to improve uptime. The results of the study were reported in a presentation by Jeff Rosen, senior industrial engineer, Aseptic Manufacturing and Packaging for Wilmington, DE-based AstraZeneca.
The presentation was made March 29 in Philadelphia at the Pharmaceutical Packaging Forum, hosted by Healthcare Packaging and Packaging World magazines.
The case study focused on an alternative analysis that included the number of downtime occurrences and equipment dependencies to identify the area with the most potential to increase uptime. The key reliability metrics used to evaluate productivity improvements that result in uptime increases and the capability to continuously run for longer intervals are Mean Time Between Failures (MTBF) and Mean Time to Repair (MTTR).
Rosen defines MTBF as the average continuous run time between the last stop condition and the next likely stop condition on a packaging line, which may be one machine or a series of machines. He says MTTR is the average time to resolve or repair the cause of the stop condition and begin running the line again.
"The focus is on the reliability and uptime connection, showing how reliability is a subset of OEE (Overall Equipment Effectiveness)," says Rosen. "A lot of companies will use OEE as a way to target where they are going to focus their improvement. We need to understand how to frame reliability within that. It will be important to see the reliability metrics so that you can use them." The remainder of this story comes from Rosen's presentation.
Defining reliability
Reliability is the measure of the likelihood that a product or system will operate without failure for a stated period of time on a packaging line. How long can we run before a system failure shuts down the line? Uptime is the time during which a piece of equipment or a packaging line is functioning or able to function. We need to target reliability to increase uptime using these metrics.
One way of doing that is to look at OEE. But we want to focus on losses during the run cycle, determining the breakdowns or minor stops.Typically when we go after uptime improvements, we look at percent uptime over a period of a shift, a week, or a month, examining minor stops in minutes. We look at speed of the line and we also look at equipment stop downtime analysis. We develop a chart showing our one biggest downtime category and usually target that area for improvement.
More detailed analysis
But we have to understand that determining the biggest downtime category doesn't give us the whole picture. We need a more complete analysis. The way we do that is to focus on the reliability losses during the run cycle. When a line goes down for 10 minutes or longer, we call it a breakdown. If a line is down for less than 10 minutes, then perhaps an operator is clearing up a jam. Those are minor, short stops.
We need to determine if we're running at a designed rate or at our maximum demonstrated production rate. Are we running at that rate whenever we are running? And are uptime losses due to the affect of upstream or downstream equipment? Typically we have more than one machine on a line, with several machines in a series, so we need to know the interaction between those machines. And what is the interaction between the faults that can occur just within one system or one machine? That gives us the full picture if you look at all these metrics. We do this because increased reliability results in less wear and tear on people and equipment.
Certainly if you are an operator relying on equipment that's constantly stopping, you have to fix it. Minor stoppages can result in bigger breakdowns. They can be expensive to fix and may result in the machine being out of service for some time. That eats into our ability to meet our production schedules and the ability to increase our output capacity if we need to. We want to improve reliability to ensure we can deliver the medicines our patients may require, so it's real important that we keep the line up as much as possible.
Measurement 'visibility'
A measurement system should be able to analyze data, identify improvement targets, and indicate if implemented improvement actions are working well. Some metrics are not analyzed in a format that makes them actionable. Actionable metrics are those that indicate the source of loss areas that with improvement will directly increase uptime or mean time between failures. By drilling down through the OEE performance metric we can gain visibility to the reliability-related output loss areas that may offer the biggest bang for the buck.
For example, we know what the OEE of the production line is. We know that the three components of OEE are availability, performance, and quality. In this case study, we're going to break performance down and look at the internal machine faults. We need to have that visibility of the internal machine faults, the minor stops, the rate loss, and also PLC signals that could tell us the conditions that may eventually cause a minor stop.
We need to look at meantime between failures to see what is the uptime interval duration. At any given point of time when we are trying to run, how long will the machine stay running before it will stop again for a system failure? Then we want to look at stop occurrences. How many stops do we get? How long do they tend to be? What are the stop reason relationships? When does a stop occur and why is that? What is the meantime to repair? How long does it take for the mechanic or the operator to recover and get the machine back up and running?
Next, we need to look at the output loss due to running at less than target speed. We really need to have the right analysis done so that we can figure which of these metrics we should hone in on to provide the biggest bang for the buck.
...Read more
By Jim Butschli, Editor, Healthcare Packaging
The presentation was made March 29 in Philadelphia at the Pharmaceutical Packaging Forum, hosted by Healthcare Packaging and Packaging World magazines.
The case study focused on an alternative analysis that included the number of downtime occurrences and equipment dependencies to identify the area with the most potential to increase uptime. The key reliability metrics used to evaluate productivity improvements that result in uptime increases and the capability to continuously run for longer intervals are Mean Time Between Failures (MTBF) and Mean Time to Repair (MTTR).
Rosen defines MTBF as the average continuous run time between the last stop condition and the next likely stop condition on a packaging line, which may be one machine or a series of machines. He says MTTR is the average time to resolve or repair the cause of the stop condition and begin running the line again.
"The focus is on the reliability and uptime connection, showing how reliability is a subset of OEE (Overall Equipment Effectiveness)," says Rosen. "A lot of companies will use OEE as a way to target where they are going to focus their improvement. We need to understand how to frame reliability within that. It will be important to see the reliability metrics so that you can use them." The remainder of this story comes from Rosen's presentation.
Defining reliability
Reliability is the measure of the likelihood that a product or system will operate without failure for a stated period of time on a packaging line. How long can we run before a system failure shuts down the line? Uptime is the time during which a piece of equipment or a packaging line is functioning or able to function. We need to target reliability to increase uptime using these metrics.
One way of doing that is to look at OEE. But we want to focus on losses during the run cycle, determining the breakdowns or minor stops.Typically when we go after uptime improvements, we look at percent uptime over a period of a shift, a week, or a month, examining minor stops in minutes. We look at speed of the line and we also look at equipment stop downtime analysis. We develop a chart showing our one biggest downtime category and usually target that area for improvement.
More detailed analysis
But we have to understand that determining the biggest downtime category doesn't give us the whole picture. We need a more complete analysis. The way we do that is to focus on the reliability losses during the run cycle. When a line goes down for 10 minutes or longer, we call it a breakdown. If a line is down for less than 10 minutes, then perhaps an operator is clearing up a jam. Those are minor, short stops.
We need to determine if we're running at a designed rate or at our maximum demonstrated production rate. Are we running at that rate whenever we are running? And are uptime losses due to the affect of upstream or downstream equipment? Typically we have more than one machine on a line, with several machines in a series, so we need to know the interaction between those machines. And what is the interaction between the faults that can occur just within one system or one machine? That gives us the full picture if you look at all these metrics. We do this because increased reliability results in less wear and tear on people and equipment.
Certainly if you are an operator relying on equipment that's constantly stopping, you have to fix it. Minor stoppages can result in bigger breakdowns. They can be expensive to fix and may result in the machine being out of service for some time. That eats into our ability to meet our production schedules and the ability to increase our output capacity if we need to. We want to improve reliability to ensure we can deliver the medicines our patients may require, so it's real important that we keep the line up as much as possible.
Measurement 'visibility'
A measurement system should be able to analyze data, identify improvement targets, and indicate if implemented improvement actions are working well. Some metrics are not analyzed in a format that makes them actionable. Actionable metrics are those that indicate the source of loss areas that with improvement will directly increase uptime or mean time between failures. By drilling down through the OEE performance metric we can gain visibility to the reliability-related output loss areas that may offer the biggest bang for the buck.
For example, we know what the OEE of the production line is. We know that the three components of OEE are availability, performance, and quality. In this case study, we're going to break performance down and look at the internal machine faults. We need to have that visibility of the internal machine faults, the minor stops, the rate loss, and also PLC signals that could tell us the conditions that may eventually cause a minor stop.
We need to look at meantime between failures to see what is the uptime interval duration. At any given point of time when we are trying to run, how long will the machine stay running before it will stop again for a system failure? Then we want to look at stop occurrences. How many stops do we get? How long do they tend to be? What are the stop reason relationships? When does a stop occur and why is that? What is the meantime to repair? How long does it take for the mechanic or the operator to recover and get the machine back up and running?
Next, we need to look at the output loss due to running at less than target speed. We really need to have the right analysis done so that we can figure which of these metrics we should hone in on to provide the biggest bang for the buck.
...Read more
By Jim Butschli, Editor, Healthcare Packaging
![Alison Bryant, senior consultant at Antea Group and communications director for the Healthcare Plastics Recycling Council (HPRC), addressed the crowd at the[PACK]out.](https://img.healthcarepackaging.com/files/base/pmmi/all/image/2024/05/Alison_Bryant.6644005a513b0.png?auto=format%2Ccompress&fit=crop&h=167&q=70&w=250)
