Editor's Note: The following question comes courtesy of Reddit. I was on Reddit, saw this inquiry and found it interesting. Given that Larry Fast writes our Ask the Expert: Lean Leadership column, I asked him to weigh in on it—which he did in the following column. If you have a question you would like Larry to address, please use this form to submit it. We want your questions.—Jill Jusko, associate content director.
Q: “I am working as a manufacturing manager in a manual (batch) process plant. Our management has decided to go lean and asked me to come up with the cycle times for each step for all varieties of products. So, I observed the processes and came up with close to 150 sample times for every step. The range is high - varying from seven minutes to 45 minutes. Now in this case, how to come up with an ideal cycle time based on which our plant's productivity will be calculated? This is the case for 80% of the processes!! I am not willing to go with the least (or best) time as they can be outliers and may not reflect the reality. Is there any process/tool that can help me?”
A: Wow, you’re faced with quite a challenge as you contemplate how to respond to your management. Before we get into what to do, let me start with what not to do.
Do not use averages to come up with “an ideal cycle time.” That’s like putting your head in the refrigerator and your feet in the oven and thinking you’ll be comfortable. When you have a range of cycle times from seven to 45 minutes, you must have the ability to know what machines are capacity constrained so you can accurately schedule machine time that delivers great on-time delivery to customers. There is no “ideal cycle time” that applies across the board except on products that follow the same process routings, use the same or similar materials, et cetera. You are wise not to fall into this “average ideal cycle time” trap. You’ll be criticized at every turn.
Do not tie productivity measurements to just a machine cycle time. Other things often affect cycle times, such as having operators in training, maintenance issues, scheduling issues, raw material issues, quality issues, and more. The better metric to use is OEE (overall equipment effectiveness), which takes the variables into consideration and highlights the cause of any shortfalls or over accomplishments. Resources can be directed at the root cause of process problems that make the “ideal cycle time” impossible to deliver until corrective actions are taken.
Now let’s consider what to do.
1. "Is there any process/tool that can help me” Yes! It’s the value stream map (VSM). The objective is to understand the capacity necessary to make each product type based on customer demand in a predetermined time frame. You can define both the capacity burden and the flow by doing a VSM. (Mike Rother and John Shook’s book, Learning to See, is the gold standard for learning this process.) If you don’t have the talent in house, I’d recommend you hire a consultant/trainer to equip your team with the skills necessary. The objective is simply to balance your lines around the appropriate constraint to optimize throughput. Absent the VSM knowledge, you’ll have work-in-process stack up in excess in front of the constraint, which isn’t good for the customer or the business. Reminder: You’re looking for commonality of products and processes that have the same or very similar flow. Then balance the line, i.e., the necessary support equipment, to complete the product.
2. Be data driven and use the Pareto principle (80-20 rule) to identify and focus on the best opportunities first to improve the business. Using a cross-functional team is critical to the effort—for example, the supervisor, engineer, scheduler--and pull in other resources as required, such as machine operators, maintenance personnel. Move quickly with a keen sense of purpose.
3. Train up your team, for example, on the Theory of Constraints, VSM, and your operators to understand and participate in doing the VSM work. Operators who understand this process and the importance of keeping constrained work centers running should have a loud voice at the table. They are closest to the process and deserve the respect of everyone else who is involved. Be proactive with operator education and training or you will struggle to make the profound impact that should result. Once knowledgeable operators engage, they’ll understand why being alert for new constraints presents opportunities for ongoing improvements. Every improvement helps to create new, free in-house capacity while also deferring capital expenditures into future periods. Kaizen events are certainly a great way to get the ball rolling. But education comes first.
4. Make new crewing models. In a traditional batch operation, operators are assigned a machine or group of machines to run. In a value stream, the constraint should run all the time except for necessary changeovers and preventive maintenance. This requires a more flexible group of operators.
4a. The first priority is to have multiple people on each shift capable of running the constraints. When an operator calls in sick, the constraint still needs to be run. A person from a non-constrained machine will immediately move to the constraint at shift change and the supervisor will help the remaining crew flex to supply the necessary materials to support the rest of the shift. (In larger operations, “floating” operators are on staff 24/7 to cover absences based on historical records of absenteeism. They usually are the ones who train new operators because they can run any machine in the area assigned.)
4b. The second priority is to cross train all operators to run multiple machines used in their area to always ensure coverage and provide a steady flow of work-in-process supply to the constraints. The guideline we used at a former company was this: Other than the assigned machine, an operator needed to learn how to run two other machines in the value stream within a certain period. Incremental pay increases were baked in to recognize the new skill and flexibility these operators had accomplished.
In summary, it’s relatively easy to get a good read on where cycle time improvements will immediately, positively, impact the business. Over time, you continue to improve cycle times on the constraint until the constraint moves to a different machine and then your efforts go to work to move the new constraint. Again, and again. This is what we mean when we talk about continuous improvement!
Before closing, I want to comment on the direction coming from senior management in this, or any other, business. The “boss has decided to go lean.” That’s often very scary to hear, knowing that the boss doesn’t understand the words he just said. "Going lean” is based on what expectations? What mindset? Companywide or just in the shop? Have the boss and the board of directors studied the topic and made a career-long commitment to excellence themselves? Is it in the strategic document as a high-priority forever? Has necessary education and training taken place, or is it being prepared to get started? If so, great and go get 'em.
Unfortunately, I’m concerned that may not be what’s happening here. I’ve never seen a CI initiative, by any name, ever start with an assignment to specifically go develop an “ideal cycle time” that would be widely applicable in what appears to be a job shop. My take is that it’s a boss who doesn’t understand CI, nor how to balance lines to support constraints. My fear is that the boss may be on a path to demand change—change that could result in forcing improperly engineered expectations (unrealistic and wrong-headed) on the operators. Forcing this will alienate the very people who know there’s a better way, but whose heads and hearts won’t be in it.
“There is no right way to do a wrong thing.” -- authors Ken Blanchard and Norman Peale
“Diplomacy is the art of telling people to go to hell in such a way that they ask for directions.” --Winston Churchill