Category Archives: Uncategorized

My Lean Transformation

”How many segments does an earthworm have?”

That question from the veterinary school’s entrance exam still haunts me. Who takes an earthworm to a veterinarian? I failed the test. My brain is not wired to store and retrieve data. It was little consolation at the time that I aced the reading comprehension and problem solving portions of the exam.

Ironically, my science degree led to a quality assurance position at a pet food plant. The job was to segregate questionable product out of the endless stream of product, sample, test, and report findings to management. When things were running well, which was most of the time, I had five minutes of work per hour. At first, the job was awesome. Someone was paying me well to play cards all night. However, a guy can only play five hours of pitch five days a week for a few years before it becomes monotonous. Although many would covet a job like this, boredom made this job a nightmare for me.

Many unhappy months followed; mostly spent watching the second hand on the clock. Each night dragged on and on. The lack of long term memory kept me from seeking a more technical lab job. Then, finally, at my lowest moment, I saw it!

At first glance, it appeared someone brought their yearbook to work. Never seen a catalog with a hard cover. The pages were heavier and glossier than you would expect. Four to six products were artistically displayed on a page. As with the catalog, the merchandise was high quality. The same quality level the company desired. The company would give points for each accepted suggestion that could be used to obtain merchandise.

A forklift almost hit me. For some reason, a safety mirror wasn’t added to this blind corner. My first accepted suggestion yielded 5,000 points. Very few items could be purchased with the minimum award of 5,000 points. The fishing pole I coveted was 20,000 points. I challenged myself to earn those additional points. Progress was slow. Eventually, I learned every problem irritates someone: employee, customer, supplier, etc. If I seek irritated people, I would find problems. Once a problem was identified, finding a solution was easy. A new door to facilitate access to a rooftop thermometer gave me enough points to get the best fishing rod I have ever owned. Soon, new boots and a wall clock found their way to my house.

The atta boy letters that accompanied the points award checks were too pretty to throw away. One day, I noticed I had collected quite a few of them. This is when my goal changed from obtaining merchandise to averaging one improvement per month.

This challenging goal pushed me to my limit. After five years on the job, I was finally motivated to read the voluminous quality assurance manual and the product specification manuals. I looked for trends in reams of consumer complaints. I listened to anyone with a problem. In three years, I initiated forty-two improvements. Only a couple of my suggestions were rejected. One of those rejected, using an incident journal for each machine, was a subject in one of my future MBA classes. The procedure of comparing machine data to customer feedback reveals the optimal machine settings/procedures.

Although I still collected pots and pans, dishes, utensils, a rifle, the most comfortable robe I have ever worn and toys for a children’s shelter, I would have initiated the improvements without the merchandise and did so with other companies. I learned most front line employees are open to lean.

What’s your lean transformation story?

What irritates you?

It was Ernie’s last week of work. My retirement present was to check the rooftop thermometer for him. It was a miserable experience. The access to the roof was on the other side of the plant. The roof was dark with no clear path to the thermometer. A couple times I had to cross a waist high wall. The wind was howling, the roof was slick, and the temperature was below zero. If I had slipped and couldn’t have gotten up, no one would have looked for me before frostbite set in. The last obstacle was a snow drift.

Incredibly, employees have been doing this task six times a day for years without complaining. A stairwell was present on the other side of the wall where the outside thermometer was located. A couple of steps and a door made checking the rooftop thermometer easy. Operators extended their breaks by fifteen minutes to check the rooftop thermometer. Since, a relief operator was needed for breaks and the new door made the long break unnecessary, the new door saved the company $6,000 per year in labor costs and it eliminated a safety hazard!

How could we have discovered this problem sooner?

Irritations are collected. For example, “I had to walk across the roof in a blizzard last night twice!”

Post these irritations above the suggestion box and solicit suggestions. Over four hundred people worked at the above mentioned factory. One of them would have discovered the need for another door.

Human See, Human Do

The research chemists did not know how to use the FTIR. The flawed procedure was:

  1. Search through the computer disks to find the last scan for the particular ingredient/finished product being tested. Info for sample preparation is included with the scan.
  2. Prepare and scan the new sample and save on disk.
  3. If a customer complaint is received, compare scan to control.

I told Brenda the new sample must be compared to the control sample before the product left the plant. She yelled and cursed at me. I refused to use the flawed procedure. Brenda continued her verbal attack.

The first step was to check past FTIR scans. The sample preparation was consistently different between scans of the same ingredient/finished product. One day the sample would be placed between two crystals; the next time the sample would be dried on one crystal. This made control comparison impossible.

Two changes were needed to correct the procedure. Searching disks to find the last sample took too much time. Hence, control FTIRs for all ingredients/finished products must be kept on a single disk. If more than one disk was needed, labels could identify which control disk was needed. A five minute search per sample was eliminated.

A notebook was needed to record sample preparation for each ingredient/finished product. For example, Ingredient X-71: 2 drops from a small straw spread over entire crystal before drying.

While the sample was being prepared, the control sample could be uploaded for comparison, saving more time. Since the sample is compared to the control immediately, saving the scan was not necessary. Finished product samples were kept for years, so if a customer did complain another scan could be prepared.

The proper procedure took a fraction of the time the flawed procedure took. Contaminated ingredients could be rejected before use and customers could be assured of receiving good product.

A few weeks later:

The following test was used to test the efficiency of cleaning products:

  1. A carefully soiled linoleum tile was taped to the outside bottom surface of a metal tray. This tray had an opening which exposed the top of the soiled linoleum.
  2. The tray was placed on the washability apparatus. 100 mls of diluted cleaner was poured on the linoleum. A scrub pad was placed in its holder and a weight was added to the top of this scrub pad holder. After the cleaner had been on the soiled linoleum for one minute, the machine was turned on. The machine moved the scrub pad back and forth across the soiled linoleum. After 75 strokes of the scrub pad, the cleaning was terminated and the linoleum was allowed to dry.
  3. The cleaned linoleum was placed on another instrument , which appraised the amount of dirt removed. This instrument placed a numeric value on the whiteness of the cleaned linoleum. This whiteness value allowed comparison between cleaners.

The problem with this test was the cleaner dissolved the adhesive on the tape and the cleaning solution leaked off the linoleum before the scrubbing began. Hence, when the scrubbing began, no cleaner was left on the soiled linoleum tile. I tested different types of tape to no avail. Then something amazing happened. Brenda found the answer. Four clamps, on each corner of the tray, worked perfectly to keep the cleaning solution from leaking off the linoleum. No tape was required.

There are only a few lean role models around. A good substitute is a lean story.

Lean starts with the need to achieve

Originally published at The Lean Post.

McClelland’s Theory of Needs contends each individual is motivated by a combination of three needs: achievement, power, and affiliation.

Need for achievement. People high in this need have an intense desire to succeed in meeting challenging goals. They want to get the job done quickly and well. These people may falter when a challenge is not available.

Need for power. People high in this need are eager to control the actions and behaviors of others.

Need for affiliation. People high in this need have an elevated desire to be liked and accepted. They want to work with their friends.

Visualize each employee with a cup of needs. Each person’s cup will contain all three needs to some extent. Some cups will have a large amount of one need, others will have equal amounts of all three. Some cups will have high levels of two needs and a small amount of the third. Many combinations are possible.

In a sweatshop economy, a manager with a high power need is required. In a lean environment, a leader with a high achievement need is mandatory. A sweatshop manager does not want to work with friends. A kaizen leader does, so she will have a good chunk of the affiliation need in her cup.

Here is an example of the interaction between the need to achieve and the need to exert power in an organization:

All the lab techs told lab management the pH meter was not giving accurate results. Lab management’s only response was, “It is the responsibility of each lab technician to calibrate the pH meter before use.” So most lab techs began adding extra citric acid to one of the products to get the pH in spec.

The lab supervisor and manager then assumed all the production techs had suddenly started weighing the citric acid incorrectly. They were sure all of their lab instruments worked flawlessly. A lab tech suggested that the team make a control sample in the lab to compare the production sample to and to keep each production sample to determine if the pH changed after the meter was recalibrated. The lab manager rejected these ideas. The only acceptable fix was for each lab tech to calibrate the pH meter before use.

Finally, a temporary lab tech called the pH meter manufacturer and asked for help. He learned that the probe should be washed with soap and water after every use. After the lab techs began doing this, the pH meter worked flawlessly again. No extra acid was needed.

When the plant manager became aware of the situation, lab management claimed they were never informed of any problems with the pH meter.

As you can imagine, the lab manager and supervisor had high power needs. They just wanted the job to get done and little to no interest in real problem-solving. The lab tech who called the pH meter manufacturer and the lab tech who wanted to use control samples had high achievement needs. They wanted to do the task correctly. They wanted to address the problem.

Soon after this particular incident this company embarked on their lean journey. The power brokers led the initiative, which meant the company’s transformation never really had a chance. Those with a high power need will insist again and again that everything is fine, no changes are necessary. They don’t like employees who rock the boat.

So in my experience, the first step in a lean implementation is to identify employees with an intense desire to achieve. Find those people who are already practicing lean, whether they know it or not, by being willing to see and engage with problems. To find people with a high achievement need, ask your employees for a list of prior achievements and suggestions. These are your MVPs and future kaizen leaders. They are your ticket to a successful lean journey.

The “Who was helpful?” Culture

This sanitation procedure is unsafe. Why isn’t the new plant manger correcting them? The meeting is almost over. Say something! People are standing up. I do not want to be a know-it-all at my first training session. I must do this.

“This procedure is not correct. You put a tag on a sanitized tote to let everyone know the tote can be used for rework. When an employee starts using this tote, he takes the sanitation tag off the clean tote. Quality assurance then has a few hours to retagged the tote as safe to use.

When the floor drain clogs and totes are moved around, errors are likely to occur. Which tote was sanitized? Which tote needs to be sanitized? Nothing is tagged. The sanitation tag must be left on the rework tote until QA adds its identification.”

The plant manager quickly seconded the motion. The procedure was reluctantly changed.

Things were not going well before this training session. First shift treated me like I was an idiot. After the training session, things got worse. First shift refused to talk to me. The person I relieved did not update me on critical factors. He told the other second shift quality assurance tech. I was being put in my place. I finally told the other second shift tech that he would have to handle any problem on my side if he continued to support this blackballing.

Every day, third shift employees left notes in my drawer or on the bulletin board. “I would thank you for helping us but since you didn’t, I won’t.” Third shift’s first task was to take samples from newly filled storage tanks. Daily records revealed they didn’t take samples until two or three hours after starting their shift. I ran eight hours everyday. If anything, they should be helping second shift. When I mentioned this, the notes stopped.

When I was eligible for my post training raise, the boss was out of town. I asked the human resource manager when I would be getting my raise. I will always remember her exact words, “Last I heard, you were not doing well.” I was breathing fire when the boss returned. I received a raise without saying a word. Later I found out it was only a fraction of the normal raise.

The plant manager never talked to me again.

When the next employee started in the lab, someone asked if I would help make her look incompetent. Otherwise, she would get the next promotion.

When the plant held its one and only kaizen blitz, the guy who asked me to sabotage a newbie’s career was the lab’s representative. The blitz’s goal was to identify oil waste. The main waste of oil was actually not using the proper amount of water in margarine. Less water used equates to wasted oil. I did not help. The blitz only focused on oil going down the drain. A minuscule amount compared to the 2.5% waste of oil in the margarines.

How could we have caught this problem sooner?

The problem was competent employees were being forced out. At least six tenured, trusted employees against the newcomer. It was easy for management to side with the majority.

I prefer to make my own judgement on the competence of each individual. My first task is to objectively identify innovative behaviors. Correcting a procedure during a training session would be an innovative behavior. Anytime a procedure is changed, I want to know. Achievement is a key component of conscientiousness. People who care about our customers make improvements.

Employees exhibiting innovative behaviors will be asked, “Who was helpful?” Only helpful people will be promoted.

The simplest and fastest way to identify innovative behaviors is to ask each employee for a list of lean stories.

Leroy’s Lean Story

On my second job, I investigated customer complaints for a floor care products manufacturer. One summer, paint was peeling off gym floors at an abnormally high rate. Lab work is basically comparing bad product to good product. Comparing the bottom of paint peelings to the bottom of paint still adhering to the floor was impossible. This problem did not appear to be solvable.

Finally, I noticed red paint never peeled off customers’ gym floors. The lab’s paint adhesion test was inaccurate. Red paint did not always adhere to wood in the lab. I needed a substrate that would give accurate and consistent results. I applied red paint to everything I could get my hands on: steel, glass, ceramic tile, aluminum, concrete, etc.

Red paint always adhered to glass. The colors that peeled did not adhere. Leroy, another research chemist, used my adhesion test to add the necessary amount of bonding agent to the paints that peeled. New formulas were sent to our paint manufacturers. My paint adhesion test became a Quality Assurance procedure and was shared with suppliers. The company’s most pressing problem vanished.

A few weeks later, Quality Assurance was performing my test when I walked through the door. They were puzzled when I kept peeking over their shoulders. When I told them that was my procedure, they all laughed. The new paint adhesion test was Leroy’s idea

SOP changes must be shared

This is the answer to the question “How could we have caught the problem sooner?” to last week’s story, My favorite lean interview question.

Vicki was taken into the lab manager’s office and interrogated by a lab tech chasing the promotion carrot:

  • What are the Lab Rats hiding?
  • It’s best for your career if you come clean now.
  • Why would the Lab Rats do something like this?

These questions could not be answered. The Lab Rats weren’t hiding anything. Each week, the lab manager received a copy of our meeting minutes. That was our team’s rule, not the plant’s standard operating procedure (SOP). When Vicki was queried, the lab rats were between projects. The turkey project was over; we had been producing good turkey products for weeks.

After an hour, Vicki left the interrogation room and headed for the parking lot. Thankfully, the lab manager caught up with her and asked her to come back to work. The manager would not tell me why this happened.

Weeks later I learned the Quality Assurance manager had taught the low protein supplier how to check the moisture content of the ground turkey before it was shipped to our plant. That was a good idea. If the turkey was out of spec at our plant, the turkey would have to be reloaded on a semi and sent back. Best to make sure it is good before it is unloaded.

I do not know why this had anything to do with The Remarkable Lab Rats.

When Vicki was cross-examined, Nick, Duane and I were working second shift. We were relieved the long week was almost over. This relief soon turned to mischief. The conversation drifted towards a practical joke.

Lab techs carry radios to keep in contact with production supervisors. We decided to tape down the transmitting button on a radio and hide it in the office area of the lab. Next we moved all the office chairs into one of the instrument rooms. The plan was in place just in time.

Linda was surprised when she saw us sitting in the wrong room. In a panic stricken voice, one of us asked if she saw the Quality Assurance manager in the parking lot. After a negative response, we quietly showed her the rigged walkie-talkie. She took the bait. Linda believed the boss bugged the office and was listening to our conversations.

Vicki would be at work in thirty minutes. We had that much time to convince Linda to tell Vicki about the radio. Nick, Duane and I believed Vicki had been through enough this week. This would put her over the top; she will quit if she knew about the radio. Nobody wanted her to quit.

The four of us left the lab to continue our debate. Everyone, except Linda, was visibly upset. Finally, Linda squared her shoulders and proclaimed Vicki had a right to know and she would tell her. Her body language told us not to get in her way.

We discretely followed Linda back to the lab. Looking through the door’s window, we could see the conversation. It was one of the funniest things I have ever seen. Vicki’s eyes and mouth were wide opened. The kind of look you see in a horror movie when the victim sees the monster. Linda was very determined. We must have watched them for five minutes before they saw us. I never laughed so hard in my life. It was contagious. Soon Linda and Vicki were laughing as hard as we were. Later, when we were all talking, Vicki said all her friends were at a party and she wanted to go there instead of work. As it turned out coming to work was more fun than going to a party.

I loved and hated my job.

How could we have caught this problem sooner?

All changes to the standard operating procedure must be shared.

My favorite lean interview question

“What made you happy on your last job?”

My answer:

“Four lab technicians and I enjoyed each other’s company during a shift change. I suggested we start getting paid for talking. We were very close when we formed our team, The Remarkable Lab Rats. We hit the ground running.

One variety of canned dog food irritated all of us. The cooking temperature had to be adjusted continually during the production run. If the cooking temperature was too low, the ground turkey would not form a loaf; the contents looked like thick soup. If the cooking temperature was too high, the loaf would look small in the can. The cooking/sterilizing process took two hours. Many times, before changes in the cooking temperature took affect, the continuous flow of product would turn good. The changes then caused another downturn. We hated this product.

Quality assurance records included the number of sample cans (out of 20) with unacceptable image for each hour of production. To our surprise, 72% of the complaints for soupiness came from hour codes that only exhibited small loaf defects, 28% came from hour codes that had no image defects, and no complaints were made from hour codes with only soupy image defects. The customers’ terminology was different from ours. Our customers called a small loaf with a lot of broth, “soupy.” This is the opposite of what we labeled this image. If the plant had initiated corrective action without fully examining consumer complaints, we would have intensified the problem.

Analytical testing on this product was done at the end of the shift. Eight cans, one from each hour of the shift, were blended together for testing. This blending could mask a problem with part of the shift’s product. We decided to individually test each hour code a complaint came from. The plant kept samples of each hour code for months.

Normally, the shift’s protein level for this product was over 11%. Every hour code that resulted in customer complaints for the past year had protein levels below 11%. Our problem was low protein. Since this product was all turkey, it was easy to find the source. One supplier’s turkey shipments dropped from an average of 14% protein to an average of 12% protein during the past nine months. Twelve per cent protein turkey was still acceptable. The other suppliers kept sending 14% protein turkey, thus causing the inconsistent production runs.

A research scientist told us an ingredient in protein, collagen, causes the turkey to set up and form a loaf. As long as the protein level is high enough, all soft loafs will solidify. We already knew this minimum protein level was 11%. This had great implications for our plant. As long as we kept the protein level up, we could run on the soft side and eliminate the small-loafed, overcooked product our customers were dissatisfied with.

Our greatest achievement wasn’t turning our worst product into our best product. Our greatest achievement was teaching our company how to focus on the customer.”

This interview question is a practical application of Hertzberg’s Motivator Hygiene Theory and McClelland’s Theory of Needs.

What’s the opposite of the current process?

The boss had to show me the procedure more than once. When the product loses its yellow color, run analysis on a jug of the uncolored product. If the density is low, the product was agitated too much (whipped). Determine how much salt and color is needed to “fix” the product still in the storage tank. Complete the production run.

This did not make any sense. If the product was agitated too much, what happened to the salt and color? It should be well mixed.

The next day, before filling jugs, I climbed the massive storage tank and looked inside. The product was separated. No color could be seen. Beach ball size globules floated in a colorless oil. It did not look appetizing. The heavy ingredients like salt and color must have settled to the bottom of the tank. Lighter ingredients floated to the top. A jug from the middle of the production run was less dense because the heavier ingredients were in the jugs from the beginning of the production run. The problem was not over agitation; it was under agitation.

The storage tank had an intermittent agitation setting that kept the product stable before packaging with minimum agitation. Whoever set up the machinery must have known the product would have a tendency to separate. That’s why the jug said, “Shake well before using.” Since bad batches were incorrectly diagnosed as too much agitation, it became taboo to use the agitator. Some batches sat a day without agitation before packaging.

Our shift started using the agitator and the bad product went away.

“What’s the opposite of the current process?” is a great question. It works more than you would expect.

How could we have caught this problem sooner?

Make it easier to look inside the finished product storage tanks. The top of these tanks were close to an overhead walkway. Strolling down a ramp is much easier than climbing an oily, two-story ladder wearing a safety harness.

When a batch changes during packaging, analyze samples throughout the production run. This would have identified the high salt content in the first jugs. No one ever questioned the quality of yellow product. A batch should be consistent throughout the production run.

The Last Lean Question

The most difficult part of lean is identifying a problem to work on. Hence, the last question in a lean improvement should be, “How could we have discovered this problem sooner?”

A lean correction from the food industry illustrates how critical this last question could be. The new oil supplier did not add preservative. The parts per million (ppm) calculation was difficult. When I gave the poundage of preservative needed to the oil tank farmer, he laughed. He would need a fork truck to add that much preservative!

I double checked my math with another type of oil that used the same amount of preservative. The computerized calculation was only one-tenth of the preservative needed. Incorrect math meant unsafe products left this plant for years. Extensive internal and external audits did not identify the problem.

How could we have discovered this problem sooner?

The company had several identical plants. A oil tank farmer from another plant would have identified the miscalculation immediately. How many problems could someone from another, similar gemba easily discover? Was this problem the only reason the plant’s costs were lower than the other plants? Is the lowest cost a reason to celebrate or a reason to worry? Should all kaizen events include someone from a similar plant? Would both plants benefit? Who should we send? The last question usually leads to many questions.

07/07/2017 Before you hire someone required to do math, give them a math test.