It’s an age-old scenario manufacturers face time and time again.
Plant Manager: “We had a good week last week. What did we do?”
Supervisor: “We aren’t sure. Luck of the draw?”
How many times have you had a similar conversation? You don’t really know why the numbers were so great, and if asked to repeat it, well, you don’t think you could.
Stabilizing Processes is More Important Than Improving Them
We talk so much about improving processes and seeing all these big gains, but how can you improve something if it’s different every time? Stabilizing processes is, dare we say, more important than improving them.
One day, you calculated cycle time at 1 minute, and the next it’s 1:45 minutes. It may even vary by shift. In the first shift, the cycle time is 2 minutes and the next, it’s 2:15. The next time it takes 3 minutes. There’s so much variation in the process.
Yeah, you have a target. Most manufacturers do. You know you want the cycle time to be 1:30 minutes, every time. But, if it’s constantly changing between shifts, days, and even weeks, you’re constantly fighting against different potential issues.
How do you improve if there’s a new problem every day? If it’s not consistent, we would think it would make it pretty much impossible to find what to work on at all. How could you identify the problem? How could you create a plan of corrective action?
Paul Dunlop, Founder of Dunlop Consultants, guest on an episode of the Zen and the Art of Manufacturing podcast, and a world-renowned lean expert, explains the concept of target conditions and processes.
“The target is a mechanism to help us have an indicator around the health of our process. The outcome is telling us how healthy process is. Is it working? Is it not? Do we have some problems? We can get a bit too caught up in chasing that number and feeling bad about the reds and the misses, and we don’t have the conversation enough about the process itself. If it’s green and we’re hitting the target, then we need to be asking questions around why? Do we understand why we’re hitting the target? The stars aligned and we just had a great day? What were the factors that helped us and how do we continue to do more of that? How do we capture this? Is it documented anywhere? I think too often we skip that conversation and we focus too much on what went wrong.”
If your numbers are moving all over the place, where is the variation coming from?
Sticking with the cycle time example, maybe on the first day, there’s a piece in the machine that’s coming loose and the operator is having to take time to adjust it and restart the machine. But, then the next day, this isn’t really a problem. Instead, it’s another problem. Then, the next day, it’s yet another problem.
If you looked at only the first day, you might think, “Well, it’s a work-holding problem.” But, if you look at the other two days, there are two other problems. How do you know what to work on?
You don’t. Not really.
The idea of stabilizing processes ties into the idea of standard work. If you don’t have standard work or a standard process that you go through, how can improve it? If it’s done differently every single time, how can you improve it?
Consistency is key.
Think about those 3 different problems you’re facing, all tied back to fluctuating cycle times. Maybe you finally get it consistent and the cycle time isn’t fluctuating so widely. In this case, you may find that those 3 initial problems aren’t really the big problems, and instead, it’s something else. Consistency gives you the ability to find the real problems.
Stabilizing processes and standard work really aren’t that different in the scheme of things. They’re very connected in the world of manufacturing.
The Lean Thinker: Thoughts and Insights from the Shop Floor clarifies the concept of fluctuations and process stabilization, “You want to know enough about what is going on that you can tell the normal, intended, pattern from one which has been disrupted by an obstacle or issue.
In other words, you know how the process should operate if people don’t encounter any serious issues. This allows you to standardize.
“Once you know how the process is supposed to work, and what results you are supposed to get, you can clearly identify those times when it doesn’t work that way, or the results are not what you wanted. Now you can begin marching up the PDCA ramp toward the standard as you work to stabilize the process by systematically eliminating the issues. So Stabilize and Standardize play off one another in successive PDCA cycles.”
Think about the concept of shifts on the floor. Maybe, first shift performs a process one way and does steps 1, 2, and 3 in that order, but the second shift performs it another way by doing steps 2 and 3 before 1. You may end up with the same product, but if there are improvements needed, how can you improve a process that isn’t consistent and doesn’t follow the same standard work? How do you know what to work on?
Okay, we may be getting lost in the weeds here, but hopefully, you’re getting the point. Consistency and standard work are crucial to stabilizing processes and making improvements. That’s the most important thing to remember.
How Do You Know What to Improve?
We believe examples are the best way to educate and help people understand our thinking. This section dives further into the 3 step process examples and also gives you steps to take if you are thinking, “Yeah, my manufacturing processes aren’t very stable, but how do I stabilize them?”
In this 3 step process, each step should take ten seconds for a total cycle time of 30 seconds.
What actually happens is that the first one fluctuates. It took 12 seconds. The second fluctuates, too. It actually took 15 seconds. No surprise, the third step fluctuated, too, and took 14 seconds. In this case of a three-step process, your cycle time was actually 41 seconds, not the goal of 30 seconds.
This varies every time the process occurs, too. So, in the end, you could be anywhere from 20 to 90 seconds depending on how much fluctuation you had in the prior steps.
You are fluctuating from 30 seconds to 90 seconds. You have absolutely no predictability at all when a part is going to come out of the machine. So, it makes it very hard to plan. Maybe that part has to go into another step. Maybe, it’s finished and it’s going out the door. But, if one day you’re running at all 90 second cycle times (the worst-case scenario, but it does happen!), you’re only going to deliver 33% of what you normally would, based on the standard of that process.
But, then another day, you’re meeting the 30 second cycle time goal. Now, you’re able to meet demand and deliver on time. Management is congratulating you and things are going great. “You did a great job. Do it again next week. How did you do it? What’s going on? What do we need to fix?” But, then 2 days later, it’s back up to 90 seconds and the problem persists. When a process is unstable, it prevents knowing why something is going good, or bad, and it prevents improvements from being made.
In this example, how do you know which step is the problem? If during one of the runs, it actually did meet the goal of a total cycle time of 30 seconds, how do you know what actually went right?
It’s different every time. So, if you were going to improve something and make sure you improve the entire process, what’s the first thing you need to do? You don’t know.
You don’t know where to start so you’re going to have to figure it out. How can you consistently get 10 seconds at each step of the three-step process? It’s time for a plan of action and stabilize your manufacturing processes. There are massive benefits in stabilizing the process. Not only can you figure out where the problems are and improve the process, but you’ve actually indirectly improved it because now, it’s a consistent process.
How Can Process Stability Be Improved?
Stabilizing is frankly an improvement in and of itself, but there really isn’t a framework for it which is the hard part of all of this.
Although there isn’t a tried and true framework for stabilizing processes, you could start to incorporate the concept of the 5 Why’s to understand why the fluctuations are occurring. Why does step one sometimes take ten seconds and sometimes, it takes 20? What’s different? What’s going on?
Toyota Kata: Managing People for Improvement, Adaptiveness, and Superior Results, written by Mike Rother, explains the 5 questions to stabilize (and improve) processes.
- What is the target condition? (The challenge)
- What do we expect to be happening?
- What is the actual condition now?
- Is the description of the current condition measurable?
- What did we learn from the last step?
- Go and see for yourself. Do not rely on reports.
- What problems or obstacles are now preventing you from reaching the target condition? Which one are you addressing now?
- Observe the process or situation carefully.
- Focus on one problem or obstacle at a time.
- Avoid Pareto paralysis: Do not worry too much about finding the biggest problem right away. If you are moving ahead in fast cycles, you will find it soon.
- What is your next step (Start of the next PDCA cycle)
- Take only one step at a time, but do so in rapid cycles.
- The next step does not have to be the most beneficial, biggest, or most important. Most important is that you take a step.
- If next step is more analysis, what do we expect to learn?
- If next step is a countermeasure, what do we expect to happen?
- When can we go and see what we have learned from taking that step?
- As soon as possible. Today is not too soon. How about we go and take that step now? (Strive for rapid cycles!)
We can’t tell you the right answer. You’re going to have to investigate your processes on your own, but what we’re trying to stress here is not only the importance of stabilizing your processes, but also incorporating the “go and see” and 5 Why’s methodologies to get to the root cause of the problems.
It’s not about taking corrective action. It’s all about stabilizing your processes that then can lead you towards the path of eventually making improvements.
“With valid measurements, process improvements are difficult, if not impossible,” as stated in Chapter 6: Stabilizing and Improving a Process with Control Charts.
