Let’s have fun with this post. See how many things we did wrong in starting this new manufacturing facility and circle them. Hint: circle the entire post.
To this day, I feel very fortunate to have been a part of this work because of all the learning that I didn’t come to realize until years later.
I was a 21-year-old intern and had been selected by my manager to help design a brand new manufacturing facility in Mexico. There are only three people involved in this “top secret” plan. My manager, a consultant with an extensive computer simulation background and myself.
The goal was to design the first pull manufacturing facility in the company based on Demand Flow Technology (or DFT). DFT is one person’s interpretation of lean and how production lines need to be flexible enough to run every product at any time. Studying DFT would serve me well later in my career. I also gained a lot of experience in computer simulation of facilities.
I designed a spreadsheet that calculated the storage space required for every component and finished good across the facility based on production rates and size of components and product. This was an input into the simulation to help determine the size of the building.
We finished the design, ground was broken and I went back to school for a semester.
The new facility was opened a couple of weeks before I returned for another session of my internship.
(Pay attention here because this is my favorite part)
The Mexico facility was replacing a local U.S. facility. The company shut the U.S. facility down on a Friday and started up the new facility in Mexico on the following Monday. No ramp up for the new facility. It started it’s first production after the other facility was shutdown. There was no training of management or employees on what a “pull” facility meant and how it would be different. It was a “here is a new pull facility go run it like you ran other facilities.”
Within a month, there were over 115 tractor trailers on the parking lot storing components and finished goods. Inside the facility, finished goods were piled in any opening they could find. Television sets that were supposed to be stacked three high were six high and leaning over about ready fall. It was a complete disaster.
My manager and I were called to the floor. We were told our design and space requirements were wrong and we needed to go to Mexico and fix the problem.
I spent two days pouring over my calculations and could not find a single thing that was wrong. We got to the facility and spent a few days watching production, examining the inbound and outbound process and locating parts in the facility and in the parking lot. It became very clear that no management practices had been changed and the facility was operating in traditional batch push system.
We spent a month helping to change a few processes and get the inventory under a manageable control, but the overall solution from the high powered executives was to expand the building and keep operating as is. Not change the management practices and improve the processes.
I can’t understate what a disaster this was. Truly an enormous cluster. It was a few years later when I was leading a lean transformation in automotive that I realized how valuable that experience was.
- Having only three people involved in the design of a new facility, especially going from push to pull, is a very bad idea. It should be a larger collaborative effort. This will even help with buy-in a when the changes are made.
- Simulations are an incredible tool, but are useless when you simulate one set of assumptions and another is put into practice
- Absolutely no ramp up time for the new facility…Really!?! I am still speechless on this one.
- If you are switching from a push to a pull system, you have to train everyone from the plant manager down on how this is different and how to manage in the new system. This is crucial.
- There must be knowledgeable support for the entire facility when going from push to pull. Help everyone work through the kinks of the new processes and not allow them to fall back on old ways.
- Most important, when something goes wrong, learn and change to improve don’t fall back to old ways just because it is comfortable. In this case, it cost millions to expand the facility instead of learning new processes.
During the past weekend, I end up reflecting on how I have spent some summers of the past. I don’t know why. I just did for some reason. There was one summer 17 years ago that ended sticking in my mind that I thought I would share.
I was working for a consumer electronics company that had manufacturing in the U.S. and in Mexico. One fall, I was asked to help design a new manufacturing facility to be built in Mexico and they wanted it to be a Just-In-Time facility. This was my first time hearing about JIT, so I read up on the concept. Of course, 17 years ago almost all the material was about what it was and not how it worked.
The goal was to only have 2 hours of production materials at the production lines. I made a super fancy spreadsheet that showed how much square footage was needed in each area based on line speed, shelving, component size, packaging, etc…
In July, I was approached again and asked if I would spend the month in Mexico straightening out what was going on. The JIT system wasn’t working. There wasn’t enough room for everything.
My boss and I went over the spreadsheet three times before we went on our visit and verified all the calculations and formulas. It was all fine.
When we arrived the first day, we toured the plant. We where horrified. Televisions that were designed to stack 3 high were stacked 6 or 7 high. Boxes were being crushed and leaning. They looked like they could fall at any minute. Areas that were not designed for storage were stuffed and there were approximately 100 trailers in the parking lot with materials in them.
This was a brand new facility. It had only been open about 1 or 2 months. It was a disaster.
The first thing I learned was there was no ramp up period. On a Friday, one facility was closed. The following Monday this facility was opened and expected to run at full capacity. I had never seen any company do that before or since. There is always a ramp up period.
The second thing we learned and more importantly was there had been no training on JIT, what it was or how it worked. The facility was operating under old batch-n-queue mentality causing space to quickly fill up.
My manager and I were able to get the inventory under control through some strict inventory management processes and even get a more consistent delivery of materials to the assembly lines.
In the end, the company was not ready to run any differently. It was a shame. They ended up expanding the building and continued to run in a batch-n-queue manner. I believe the facility has been closed in the last 3 or 4 years.
It was my first exposure to JIT and all that it takes to run a JIT system successfully. I call it a system because it isn’t just about space and delivering parts. It is the management mentality to reduce changeovers, run in much smaller batches and solve problems. It really showed me how everything must work together.
Does anyone else have any horror stories from trying to implement a just-in-time system?
Small change vs. Large change is a debate I hear quite often within the Lean community.
The meaning of kaizen is to continuously make change for the better. Implied is to make small changes everyday and over time it will add up. Paul Akers at FastCap often talks about the 2 second kaizen.
Every improvement counts. This is small change.
The flip side of the discussion is large change. Transform the work into something new. Redesign the process, the layout, the flow. Act in a completely different way.
My opinion…they are both right and you should do both. The key is understanding what your organization needs and when.
If it is a traditional batch and queue organization (manufacturing or service), then as you start your lean transformation I would recommend large change. Create a pull system where the parts or service flow uninterrupted. Dramatically change the way you operate.
Once the large change is done, the improvement never stops. This is when you start looking for the 2 second improvements in the process. Everyday the process should be better. Keep making small changes.
This isn’t the only way to go about a lean transformation. It is just one way. If you want to be successful with your lean transformation take the time to really consider your strategy for going about the transformation.
All in all, some improvement is ALWAYS better than no improvement…small or large.
The other day I was having a conversation with a lean counterpart. Essentially, my counterpart mentioned they know all there is about a lot of the lean concepts and tools. This struck me as odd. I have been fortunate enough to have been studying lean for close to 15 years now. Even after implementing over a dozen pull systems using kanban, I have never felt like I knew everything there was to know about pull and kanban.
I asked my counterpart what lean meant to him. He replied it was about how a person thought and problem solved and to never stop learning. I tend to agree with this. I followed up by asking what never stop learning meant to him. My counterpart said it was to continue to learn new processes, tools, and ideas. This is where I only partially agreed with him.
Continuous learning is more than what he mentioned. It also means you never know everything about a topic, concept, or tool. In my example of pull systems and kanban I have to continue to learn. Would this work in this environment? What is the business strategy for product? Build-to-order? Off the shelf? Customized? Only one way to get it? How does the current business and world environment affect a pull system for this company or industry? Does a pull or kanban system even make sense for this business model? Or is it like and ER or seasonal crop growers?
I could go on and on with more questions. My point is, I may have done well (or not well in some cases) with a pull system utilizing kanban, but that is just a knowledge library to use to help me have better context of the current situation.
I never see myself as not learning even on something I have studied and implemented many times. Like anything else, if we aren’t learning and improving then we are regressing.
How do you answer the question of what continuous learning means to you?
People are enamored with kanban systems. This can be a good thing, but all too often they don’t understand kanban systems are there to help highlight make problems visual.
The first thing almost everyone jumps to is the calculation for the minimum and maximum levels for the kanban. I have seen some formulas that would make a mathematician with 3 PhDs blush. I don’t understand the need to have a complex formula. For years now, I have used what I see as a basic quick and easy formula to calculate the min and the max.
Min = Lead Time + Safety Stock
Max = Min + (Min/2)
Lead time is the time it takes from the moment the component is ordered until it is received and ready to be used.
Safety Stock is the amount of stock to hold because of something that could occur to delay the lead time. Base this on where you are getting the parts from, how often does something go wrong, etc… For example you might hold a little more safety stock for something you purchase from a company 300 miles away versus a component that is made in-house.
If the process is working smoothly, you will receive the component you ordered right as you get into the safety stock. When the minimum level is set properly, you will feel freaked out because you believe you will run out and right about that time the components will arrive. It is a weird feeling that you will adjust to, but makes you heartbeat fast the first few times until you get used to it and trust the process.
The maximum is something a friend and I completely made up several years ago. There is no reason it has to be this. I continue to use it because so far it has worked well for me over the last decade. I always round up to the nearest full day.
Min = 2 day lead time + 1 day of safety stock = 3 days
Max = 3 + (3/2) = 4.5 round up to 5 days
The only other number that is needed is the quantity of the product used per day. This is used to translate the number of days to a quantity of the component.
1 day usage equal 500 parts
Min = 3 days x 500 parts = 1500 part
Max = 5 days x 500 parts = 2500 parts
The point of the kanban min/max levels are to get you in the ballpark. It shouldn’t be an exact science because you will probably round to nearest full carton or order quantity anyway. Plus, min/max levels should NEVER stay static. They are dynamic and change.
I wold recommend on having what you might think is a little too much inventory to start. You can always adjust your kanban min/max levels down as you understand your process. If you start with too little of inventory, you will run out of parts and people will not have faith in the new process and give up early on before it has a chance to work.
Get rid of the waste in your kanban calculation and go and see your process to understand if your kanban min/max are appropriate.
A couple of years ago, I was working with a group to create a complex supermarket pull system. The supermarket was centered around a component that was manufactured inside the assembly facility and feed several assembly lines. A timed delivery system was going to be used to deliver components to the each assembly line every two hours. The deliveries were based on the orders the lines placed during the previous two hours. Only four hours worth of components would be stored at the assembly line and the rest would be stored in the centralized supermarket.
Below is a quick presentation I gave to the leadership to help them understand how the system would work. Slides 5 -15 show the specifics for this supermarket delivery system.
This is a specific example of how to use this supermarket system. However, the concept is the same no matter what supermarket you use. There should be:
- Centralized storage location for the components
- A small amount of inventory at the point of use
- A signal to replenish the point of use inventory
- A signal to replenish the supermarket inventory
- Replenishment triggers should be based on lead time to receive inventory from supplier
A well designed supermarket system can be a very powerful tool to help reduce inventory and simplify operations.
I know my last post was about the concept of kanban. It has been a concept that springboarded a lot of my learning over the years. It may have started with implementing a kanban system but ended up learning about visual management, the seasonality of the business, what the customer is demanding, and change management.
There are two important learnings that I have had when implementing a kanban system. Two that I thought I would share.
The first lesson is that when a kanban system is mentioned people jump to a conclusion that all materials will be handled in one way. All the materials will be set with a min/max. The min being the reorder point and the max being the point to fill the order to. This assumption scares people because setting everything to a min/max system would mean increasing inventory overall and holding inventory on some parts for a very long time. This is not a smart thing to do. People need to know that a system can be put into place that takes each component into consideration and does the right thing for that component.
This brings me to the second lesson. What is the appropriate way to handle each component?
So far, I have learned three ways to handle a component in a kanban system.
The first way is the typical kanban replenishment system. A minimum is set for a reorder point based on lead time and safety stock. The maximum is the highest quantity wanted on hand at one time. I have found the best time to use this is when a component is used on a nearly daily basis and in high quantities.
The second way is another typical way. The non-replenishment kanban. This is a kanban that is filled but not recirculated. I have found this to be best used when a component is needed for a very short period of time, a day or week, and then the component is not used for a long period of time.
The third way is what I call a seasonal kanban. It is a component that will be used frequently and with higher demand but only for a short period of time, a month, two, three. It is long enough that a non-replenishment kanban is not proper to use and a replenishment is too permanent. What I have done is set up the component on a replenishment kanban but when the use is winding down, I convert it to a non-replenishment. When the season is over the component has no inventory so things aren’t stored for an unnecessary amount of time.
Using a combination of these three can make for a very efficient system.
The digital age has been here for quit some time. One industry that has be changed significantly is the music industry. For over 50 years the music industry was a batch industry. Musicians released music in batches to the public in the form of albums. Then batches of albums would be manufactured and sent to stores before finally a consumer would buy a copy of the album.
The digital age has made it possible for the music industry to go to a single piece flow. The middle man or seller has done taken advantage of it. Now you can go to iTunes or Amazon or other websites, pick what songs you would like and download them one at a time.
Why haven’t the musicians taken advantage of this though. Musicians are still releasing songs in batches (albums) even though the consumer is downloading just certain song off the album from the internet. Why don’t musicians create a song and then release it and not wait for batches of songs to release together? It might allow more songs of theirs to be downloaded because a song is getting played, the fans hear it, and then buy it. When done in batches, only a couple of songs get played and the rest of the album may be heard by the fans if they buy the whole album or it may not.
Leveling the release of the songs in a single piece flow seems like it would be beneficial to the musicians. Allowing more of their songs to be played on the radio, which I would think would lead to more downloads and more revenue for the musicians.
Just a thought in a way to use the digital music age to their advantage. What are your thoughts?
Today’s guest blogger is Joe Wilson. Joe is a great lean thinker that worked for an automotive supplier for several years. Developing his lean thinking by diving into the deep end. Joe now works for Tyson Chicken working within their Industrial Engineering group. I am happy to post his writing here. Joe is a great lean thinker.
A few days ago, I had one of those random day-dreaming thoughts that spurred me to go look up something in a book that I haven’t read in a few years. When I opened the book, the bookmark was a heavily scribbled on piece of notepaper from a series of Lean classes I took. Among a handful of otherwise barely readable comments was these three words, underlined and circled: “PACE…NOT SPEED”.
Those 3 words stood out as a turning point for me in my lean education. Those words drove an understanding of what the huge pile of lean phrases and tools I had bouncing around my head really were all about.
You really can’t implement things like pull systems/kanbans/heijunka, standard work, or even 5-S programs until you can define what your customers want and when and how they want it. It is extremely difficult to determine how many people should be doing the work in your workplace or how much equipment you need to do the work if you don’t have a true understanding of how much work needs to be done. What this also means is that your distribution, purchasing, and planning/scheduling functions are absolutely critical to the success of your success in lean.
Where do you take this from here? Start by getting as deep as you can in what the market is for your plant/company’s output and what your customers need. Get as deep as you can in how your suppliers bring things to your door and how you handle those. Those things aren’t exactly the most interesting or flashy pieces of the pie to work on, but without a clear understanding of what you need to do, you can’t solve the problem of how you get there.
When an organization is focused on implementing lean, one of the first things that everyone talks about is focusing on the customer. It could be the external customer or the internal customer.
Who the external customer is seems to be clear in most cases. But when dealing with internal customers it isn’t always as straight forward. In my experiences, the person says their internal customer is the next physical process that changes the product (or service).
What if you have implemented a pull system with a supermarket in between two physical processes?
The functioning of a supermarket is a process. Product enters, waits, and then exits to the next physical process. The product does not physically change, but the supermarket is a process.
Even in lean notation we over look this. If you are familiar with Value Stream Mapping, the is a specific inventory icon for a supermarket. It does not get notated with a process box.
Now if the mindset is the supermarket is a process, then it is the customer of the physical process upstream and the supplier of the physical process downstream.
This changes how questions are asked around creating a pull system. Instead of asking, “How much does process 2 produce? Or what is does process 2 need?” , the questions should be, “How much leaves the supermarket in X timeframe? When does the supermarket tell me to produce?”