Operations · System Thinking

What Toyota’s success really teaches: concepts versus applications

Mohit Saini · 29 Oct 2020 · 4 min read

What Toyota’s success really teaches: concepts versus applications

Toyota is one of the most studied companies in the world. Thousands of companies have copied its methods: kanban cards, quick changeovers, U-shaped cells, visual boards. A few have achieved remarkable results. Many more have spent years on Lean programmes and seen modest improvements at best.

Why does the same set of tools work brilliantly in some places and disappoint in others?

Dr. Eliyahu M. Goldratt addressed this question in a paper titled Standing on the Shoulders of Giants (2008). His answer has stayed with me, because it applies far beyond manufacturing. Here is my summary of his argument, in my own words.

Two giants: Ford and Ohno

Goldratt starts with Henry Ford. Ford’s assembly lines achieved something extraordinary in the early twentieth century: a car moved from raw material to finished vehicle in a few days, when the industry norm was weeks. Ford did this by focusing relentlessly on flow. He even limited the space between workstations, so that work could not pile up. If the space was full, the upstream station had to stop.

Notice what that means. Ford deliberately accepted that some workers would sometimes be idle. He gave up local efficiency to get flow.

Ford’s system had one limitation: it worked for a single product with very little variety (famously, any colour as long as it was black). When customers wanted variety, the system struggled.

Taiichi Ohno at Toyota took Ford’s thinking further. He needed flow for many models in small quantities. His answer was the Toyota Production System. Instead of limiting space, he limited the number of containers between stations using kanban cards: a station produces only when the next station has used up a container and sent back the card.

Four concepts behind both

Goldratt argues that, behind the tools, Ford and Ohno followed the same four concepts:

  1. Improving flow is the primary objective of operations. Shorter lead times are what matter most.
  2. There must be a practical mechanism that tells operations when not to produce. Without it, people keep busy and overproduce. Ford used space; Ohno used kanban.
  3. Local efficiencies must be abolished. If every worker and machine is measured on being busy, the mechanism in point 2 will be defeated.
  4. A focusing process is needed to balance flow. Ford and Ohno continually found where flow was blocked and fixed that point first.

These are the concepts. Kanban, cells and quick changeovers are applications: the way Toyota put the concepts into practice in its own environment.

The mistake: copying the application

Here is the key insight of the paper. An application is designed for certain conditions. Toyota’s kanban works well when:

  • demand is relatively stable,
  • the routings are repetitive, and
  • the time to produce each item is short and predictable.

Toyota spent years making its environment fit those conditions, including persuading its suppliers to change.

Many companies copy the application without having those conditions. A job shop with hundreds of custom orders, a plant with highly seasonal demand, or a process with long and variable cycle times cannot run smoothly on kanban. When the application doesn’t fit, the results disappoint, and people conclude that “Lean doesn’t work here.”

The concepts are universal. The applications are not.

A different application of the same concepts

Goldratt then shows that the same four concepts can be applied differently where kanban doesn’t fit. The TOC application is Drum-Buffer-Rope, and in its simplified form:

  • Instead of limiting space or containers, it limits time: material is released to the floor only a fixed time (the buffer) before the order’s due date. This is the mechanism for not producing too early.
  • Local efficiency measures are dropped.
  • Buffer management acts as the focusing process: orders are prioritised by how much of their buffer has been consumed, and the reasons orders run late are recorded and fixed, starting with the most frequent.

This works in environments with high variety and unstable demand, exactly where kanban struggles. Goldratt illustrated it with a Japanese tool maker, whose results improved dramatically once it applied the concepts through time buffers instead of trying to force-fit kanban.

What this means for managers

I find three practical lessons in Goldratt’s argument.

  1. Before copying a best practice, ask what concept it serves and whether your environment has the conditions it needs.
  2. Judge an operation by flow, not by how busy its resources are. If your measures reward activity, no tool will give you flow.
  3. You need a mechanism to stop overproduction. Space, containers or time buffers: pick the one that fits your environment, but have one.

The real lesson of Toyota is not kanban. It is the thinking that produced kanban. Ford and Ohno were giants because they understood the concepts deeply enough to invent applications that fit their world. The rest of us can stand on their shoulders, but only if we look at what they were standing on.

This article summarises ideas from Eliyahu M. Goldratt’s paper “Standing on the Shoulders of Giants” (2008), in my own words.

Related reading

Mohit Saini

Strategy, supply chain & sales transformation consultant. About · Book a discussion