Just imagine if Ohno visited…

This is a “what if” scenario that people occasionally ponder, thinking about all the insider information they might learn. When you read the first-hand accounts, engineers and managers were extremely interested to have Ohno’s insights, yet most were intimidated by him. They would rather be an observer of his discussions than the one having to directly answer his questions.

There are a few recorded examples of people who worked with Ohno, some inside Toyota, others at various companies. After Ohno left Toyota he worked with the New Production System Research Association (NPS). NPS worked with about 40 companies from different industries. Here they proved that his system could improve productivity and profitability across a wide range of industries; from home building to food products to textiles to electrical equipment.

Today, most of our ‘lean’ literature is either about the tools or the softer people side of TPS.?The logic and sequence on the transformations is scarce; some conversations are recorded in “The Birth of Lean” about Daihatsu, very limited details on the nearly 40 companies introduced to NPS, and some insights in the 1973 TPS manual on their early experiments on organizing for flow.

After Ohno left Toyota in 1978 he was active in assisting other companies through the NPS Research Association and delivered seminars on TPS with staff for the Japanese Management Association. They published a book from the seminar materials - トヨタの現場管理: 「かんばん方式」の正しい進め方 / Toyota's on-site management: Proper way of "Kanban system" – closely matching the 1973 TPS manual materials.

After decades of practice and research in industries around the world, I sincerely doubt that Taiichi Ohno would approve of how most of us have applied TPS as lean.

He was not impressed with his own supervisors use of 5-S; calling it a ‘lining up’ competition. We have also failed with ‘kaizen events’, ‘gemba walks’, ‘7 wastes’, ‘one piece flow’, kanban, and JIT. We may have all the trappings of 'looking lean', yet missed the development of our production systems. Ohno could link all of his containment steps to Toyota's business needs.

Taiichi Ohno’s team first started sharing how to apply their production system outside of Toyota in 1970. (In the 1960's a few suppliers were trained on how to deliver via Kanban.) In the beginning it was all based on the personal experience of a handful of men. How they explained their approach varied with each supplier. It was haphazard enough that Ohno relented and allowed them to begin to put the system details down in writing. Until then, he had resisted writing it down because the system was still under development. Like their standard work, he had the expectation of a system constantly being improved.

Where would Ohno start?

It depends upon your production system maturity level and type of industry.

Opening sentence in the 1973 TPS Manual - “The basic idea of the Toyota Production System (TPS) is to sequence operations to improve productivity.” Ohno's objective was to create flow, which he called his "river system". By improving flow, he reduced production costs and the number of people required. His Pull system is a containment step on the journey towards achieving Flow.

All activities should move you towards achieving Flow.

For companies that he and his team helped, it was rarely a single visit. The improvement process was iterative, each step building on the one before; this could take several years. To better understand what Ohno was doing, you must look past the superficial tools we think of as ‘lean'. Start by questioning how each activity is fulfilling the Principles of Flow. All action would be adapting the production system to better reflect the Flow Principles.

When you tour most companies, it starts with an overview of the company. Sometimes this includes a bit of history, what is produced at this facility, and the flow sequence starting at the first steps. You may get a safety lecture first, then the tour group starts at the first steps to watch to evolving processes until completion.

For Ohno, a tour group would be a distraction, they would want some time to observe and ask questions. They might want to start in the finished goods warehouse. Don’t get confused by the evil inventory myths, that is not the reason. Your finished goods warehouse is a good indicator of the thinking behind your production system. It is an indicator of how accurate you are in estimating (guessing) the customer demand, how well you are rotating inventory, what the ratio of stock to sales, etc.

You can use the dust test to find some of your oldest stock. It is more than just looking at how much money is tied up in inventory. It highlights gaps in sales and marketing’s ability to gather actionable feedback from their customers (what they want, what they will buy). Even if they have stable sales, how large is their inventory of the items to sales relative to the time required to produce? If the production cycle is a few days, then why do you have months of inventory? Does the market demand fluctuate this much, or just the intermediate warehouse ordering??

Flow Principle #4: Balance Demand Pace – Marketing should understand what customers want (what they will buy), Sales should understand the market demand and smooth the variation from the market into the factory. Planning smooths production scheduling to have even loading of the production system. Their job is to supply orders for balanced loading of the production system. Even daily demand is the ideal.

Continuing the factory tour... Walking upstream to the flow can give you a rough idea of just how large the WIP (Work In Process) is. You get to see where the bottlenecks are, as well and the traffic problems moving all the WIP around. Do they have ‘intermediate warehouses’ or accumulating stock on the floor before moving it to the next operation? Look for subassembly operations that are disconnected from the main production flow. Notice rework or sorting; ask about yields. You are looking for issues (barriers to flow).

Flow Principle #1: Put Processes in Sequence – The ideal is for the materials to move directly from process to process with the least amount of transport and least amount of delay. This happens when the processes are placed in sequence, without segregating the types of processes. Secondary processes are ideally integrated into the lines to feed as needed to the next process. The goal is each process should be able to directly feed into the next process in the sequence.

Flow Principle #2: Synchronize Processes – Each process in the sequence should produce at the same pace so the production process has no bottlenecks. Synchronization requires that all processes produce 100% good quality.

• Containment until you can develop a robust process to make 100% good quality is to produce more… faster… enough to cover the losses from defects and rework.?These are processes you want to fix first. Without a robust process, you cannot create flow.

Most factories were not organized for flow, and even those well organized when new will change over time as technology and products change. It is not uncommon to find a large unused piece of equipment in the middle of a shop. It may have been replaced by a new technology, or the new product does not need this step, or the machine became too difficult to maintain.

One of the first steps is to see the flow sequence, including all the secondary processes, and identify the primary barriers to flow. This could be layout, quality problems (yields), capacity, etc. What ONE thing can you change that would improve your production system?

It is tempting to come into a new facility and give advice about moving things around. When we read the original Toyota manual, they talk about it taking at least a half a day of observing to really see what is going on. A few glaring barriers might be obvious, but the best solutions take a deeper understanding of all the factors than can be gathered in a quick walk around.

The best results happen when you invest in coaching your people.

Grow People to Grow your Production System

First Grow People

When people in leadership positions are aware of Flow and are coached with the same basic skills that Ohno started with (TWI), the results can be amazing. Teach your leaders a skill they can apply daily to solve problems.

Coach your leaders to apply Job Instruction to define exactly what must be done on critical processes to improve quality (and safety). The focus on quality can pay off in productivity improvements. The quality records may show a low defect rate, yet not be accurate in how much time is wasted working around the issue.

Would you like to have results like this?

• A company producing a commodity product (lighting) focused on growing their leadership team through practical coaching of the TWI skills and TPS. In 2016 they produced just over 12 million products, this year (2019) they will make about 48 million. With only a minor increase in staff and in the same factory. They continue to develop their staff and improve their production system. (Update: 2020 production was 60 million, even with the pandemic restrictions.)

How it was done...

• Commitment at the top to grow leaders. Support was provided, including time to practice.
• Volunteer groups of leaders to work on pilot projects in their areas (5-8 people), Awareness developed linking flow principles + TWI to the development of lean. Initial training was less than two hours.

DISCOVER FLOW - Looking at their production system from a new perspective.

• List Activities: The first skill in TWI's Job Instruction is building a timetable of training needs. To do this requires you to assess the work area and identify key areas that need improvement. To build awareness, the leaders listed all the activities in sequence, recorded relevant information on each one (issues observed, cycle times, scrap, rework, proximity between activities, WIP, etc.)
• Evaluate Options: The first part of this step was a strategy decision. Choosing your approach to selecting issues to work on. For most companies, starting with the "low hanging fruit" is most effective for learning the improvement patterns and making progress at the same time. With experience, concentrating on improving the 'system constraint' has the biggest effect of the overall performance of a facility. Other options include focusing on quality or productivity problems, and starting at the end of the flow sequence and removing barriers as you work your way upstream. Connected with choosing the problem solving strategy is goal setting. Be sure that the goals selected align with business needs. There is always the temptation to focus on cost reduction or speeding up a process. Consider both of these as outcomes to improving quality, reliability of a process, building skills of the operators and making the job easier.
• Understand Problem: Once they selected a problem, they resisted the assumption that because of their experience they already knew the answers. Job Instruction has an observation process to define the steps of a task and the key points. The key points are details that must be done to get the desired outcomes. Defining what should be happening allows you to compare and identify gaps. When they could identify an undesirable outcome, the next step was to understand why and how it happened. For this they used 5-Whys + FOG (fact/opinion/guess). The addition of FOG was to validate each answer to a question with verifiable facts before asking the next level question.
• Review Facts: With factual information available, it is easier to make good choices for the next step. An effective 5-Whys should define the root cause for an issue. Without knowing a root cause, most containment ideas are just guessing. The choice for the next step followed a simple rule of building structure first before resorting to more complex solutions. First Question: "Do you have a process?" - do you have a written job breakdown defining how to do the job. The first level for standardization was setting clear expectations on critical tasks and training everyone that would do the task. They applied Toyota's improvement sequence - people (training to standard), materials (quality issues), process (challenge to improve), machine, then design. Each level takes longer to complete and is more complex... start at the beginning.

DEFINE STANDARDS - What are the expectations...

• Create Standard: Like most companies that have been doing lean, they had 'standard work instructions' posted at the line. Most were written by engineers or quality inspectors. They did not reflect 'how' to do the job safely, correctly, and easily. Job Instruction was used on selected tasks that were problematic. We did not try to write a job breakdown on every task. The job breakdown process has a method of observing the process as a team at the place where the work is being done. The supervisor becomes the facilitator and has selected two of the most skilled operators to help write the job breakdown. Defining the steps are easy, it's the defining of the key points where the hidden knowledge of the skilled operators begins to be seen. No matter how much care you take in creating the standard, expect to miss a few key points... these missed key points will be seen as quality and safety issues.
• Validate Standard: Once the team thinks they had a complete job breakdown sheet (JBS), then the validation step starts. The three people play the roles of 'learner, trainer and observer' to test the JBS. Almost always missing details are discovered at this phase and they get a feel on how much of the JBS should be taught at a time. In the beginning, most people tried to teach too much at a time. This practice uncovers how to break the training into bite sized coaching units.
• Plan Training: Once they had a validated JBS, they would then decide who to train and define the time for training. They also used Job Instruction to define changeovers and the setting/calibration of equipment.
• Train to Standard: Since most of their workforce were experienced, the approach is different than training new hires. The experienced person usually needs only 'single point' lessons related to a process change or quality issue. For this, the training would focus usually on a singe step and the key points needed to meet their quality expectations.

CHALLENGE STANDARD - There were a few missteps of immediately changing processes or adding automation to solve problems. And the misconception of using Kaizen to reorganize the whole department.

• Understand Standard: To effectively improve any process, you first must clearly understand it. When a process was encountered that was obviously in need of improvement, the first question asked was "Do you have a JBS?". Even if we had a process with a high defect rate or its productivity was such that it was the bottleneck, we still did the observation sequence to understand the key points needed to get the desired outcomes. This actually saves time problem solving and seeing where to focus your efforts. The Job Methods process has an evaluation sequence to help you make sure you really understand the process and don't overlook a critical detail. There are five areas to 'understand'. Purpose: What is achieved? and Why is it necessary? Place: Where is it done? and Why there? Sequence: When is it done? and Why then? Person: Who does it? and Why that person? Means: How is it done? and Why that way?
• Challenge Details: Once they had the details, usually only a small segment of the task was found that would make the biggest difference. They could start with a single step, or a few... to make the biggest impact in the shortest period of time. The key question... and the first one that must be asked is "Is it necessary?" The fastest improvement could be to just stop doing a step. If it was necessary, the challenge is to find a better way. Following the same five areas in the understanding step, you would challenge to find a better way. (What could also be done? Where else could it be done? When else could it be done? Who else could do it? How else could it be done?) Commonalities were noticed in improvement ideas, so these were shared and applied in other departments.
• Develop Ideas: The hardest barrier to overcome was to develop multiple ideas for solving a problem. When there was just a single idea, the struggle became on how to make it work instead of finding the best solution. (Don't fall in love with the first idea.)
• Rank Ideas: When multiple ways of solving a problem were developed, the process became one of evaluating the pros and cons for each one. What was best for the team, best for the department, best for the company. When multiple ideas began to be evaluated together, new ideas began to emerge... hybrid combinations of multiple ideas, maybe even with a new twist.

IMPLEMENT IMPROVEMENT - There was a trap in wanting to rush off and implement ideas, buy equipment, reorganize the lines, etc... Lessons learned in being more deliberate in implementing improvements.

• Design Experiments: No matter how much experience any of us had, most ideas were just guessing based on similar experience. There were always a few unknown areas that needed to be tested before we invested time and money. The bigger the investment, the more critical it is to validate the idea. Find the weakest assumption and design an experiment: "If I do _____, then I expect _____ outcome."
• Evaluate Results: Some experiments were simply moving a table or box a little closer to the operator. We could get immediate feedback to see it it worked. If not, we returned it to the original location and sought to understand why the actual results were different. It was only when things DID NOT WORK AS PLANNED that we learned and improved our skills. The primary question for an improvement is "Did we meet our objective?" If not, we went back to improve again.
• Plan Improvement: The complexity of an improvement was a factor in how detailed the implementation plans needed to be. Very complex included creating custom automation or redesign of the product (DFMA). Less complex were quality improvements that required retraining staff, or detailing how to calibrate machines, or moving a few work benches. Sometimes the improvement only reduced a defect rate 50%. These were regularly used, and we returned to look for more clues and better ideas.
• Follow Up: There are two parts to this step. First is to make sure that after the initial excitement of implementing a change, the improvement really worked and people continued to follow the improved methods. The second part of follow up was the coaching of the leaders to support their following the improvement patterns. The Follow Up Assistant can help train people to apply the patterns (skills)... better to teach the use of the skill than to give solutions.

Reading materials:

My Favorites: The Birth of Lean; Management Lessons from Taiichi Ohno; Jon Miller’s translation of Ohno’s Workplace Management, and the 1973 TPS Manual.

• The Birth of Lean: Conversations with Taiichi Ohno, Eiji Toyoda, and other figures who shaped Toyota management – edited by Koichi Shimokawa and Takahiro Fujimoto. Translated by Brian Miller and John Shook
• Management Lessons from Taiichi Ohno: What Every Leader Can Learn from the Man who Invented the Toyota Production System – Takehiko Harada
• Taiichi Ohnos Workplace Management: Special 100th Birthday Edition – Taiichi Ohno, Jon Miller (translator)
• トヨタの現場管理 : 「かんばん方式」の正しい進め方 / Toyota's on-site management: Proper way of "Kanban system" - Japan Management Association (Nihon Nōritsu Kyōkai)
• Copy of TPS manual available by request. ([email protected])
• Toyota Production System: Beyond Large-Scale Production - Taiichi Ohno
• Just-In-Time for Today and Tomorrow - Taiichi Ohno
• Workplace Management – Taiichi Ohno
• NPS New Production System: JIT Crossing Industry Boundaries – Isao Shinohara, Editor
• The Evolution of a Manufacturing System at Toyota - Takahiro Fujimoto

For Serious Researchers - Flow

"Fordism Transformed" - Chapter 1: Emergence of the 'Flow Production' Method in Japan. (Kazuo Wada) This chapter details some of the early experiments of applying flow production in the aircraft industry and documents the transfer of this knowledge into Toyota. Toyota's experiments on installing 'flow production' started with their Koromo Plant. One of the interesting things they did was to mount some of the equipment on 'duckboards' so they could easily rearrange the sequence of machines.

"My Forty Years with Ford" - Autobiography of Charles Sorensen. He was Ford's 'right-hand man' for most of his career. He writes about Walter Flanders reorganizing Ford's factory for flow in 1906. Production jumped from 1600 to 10,000 in the first 12 months under Flander's direction. Sorensen also describes laying out the Willow Run Plant (aircraft) based on the same ideas he learned from Flanders.

"Ford Methods and the Ford Shops" - Horace Lucien Arnold and Fay Leone Faurote, 1919. This is a collection of articles detailing the production activities in the Highland Park facility starting about the time they introduced the assembly line. The thinking of balancing the workloads went far beyond the assembly line. Shift changes were staggered over a two hour period to level the load on transportation to and from the factory. Every day was a payday for the payroll department. Before the computers, all payment was done manually. To level the load on the office, they had a rolling schedule of payroll depending upon your department.

"Ford Men and Methods" - Edwin Norwood, 1931. This has a chapter describing what we now describe as an 'andon' cord system. It details that the men were instructed to stop the line whenever they encountered a problem. The had a central board with lights to identify which station needed help. If the light was on more than two cycles, the manager and maintenance were dispatched.