TOC in Operations – Advanced · Lesson 1 of 12
The standard solution in one session
A fast recap of TOC in operations: the flow problems, the conflict behind them, choking the release, buffer management, due dates from the load on the constraint and improvement from the reds. Then the nine assumptions the standard solution quietly rests on, which the rest of the course breaks one at a time.
This course starts where TOC in Operations – Basic ends. If you have implemented Drum-Buffer-Rope, you know it works in most plants within weeks. You also know the plants where it didn't: where on-time delivery climbed to 80% and stuck, where the constraint kept moving, where one order broke the whole system. This course is about those plants.
First, the standard solution in one page.
Why on-time delivery is poor
Ask a plant why it delivers late and you will hear: capacity is short, people are not efficient, vendors are unreliable. Then look at the facts. A typical order takes about 50 days to go through a plant where the work on it, the touch time, is five or six days. Some items of an order reach the warehouse early while others are late. Dispatches pile up at the end of the month.
That is not a capacity problem. It is a flow problem, and it has four forms:
| Flow problem | What happens |
|---|---|
| Over-activation | Resources make items nobody needs now |
| Resource stealing | Capacity goes to some items at the cost of others |
| Material stealing | Common material goes into items that aren’t needed yet |
| De-synchronisation | A long gap between the first and the last part of an order arriving |
The shape of the plant decides which ones dominate: V plants (few materials, many end products) suffer stealing; A plants (many parts, few products) suffer de-synchronisation; T plants suffer both; I plants are mostly simple.
The conflict behind them
Every plant manager wants to manage the shop floor well. To control cost, he must improve local efficiencies: batch, keep everyone busy. To ensure flow, he must complete orders fast, which means not batching. Underneath the efficiency side sits one assumption: an idle resource is a waste. It is true only at the constraint. Everywhere else, a resource must have more capacity than the constraint, or it cannot rebuild the constraint’s work bank after a disruption, so it must be idle part of the time.
The four mechanisms
1. Choke the release. Material is released only a buffer-time before the due date. Because touch time is under 10% of lead time, the buffer can start at half of today’s lead time. Less work in process means less waiting, so lead time falls without any loss of output.
2. Buffer management. Each order’s colour shows how much of its buffer is used: green under a third, yellow to two thirds, red beyond, black when late. The floor works blacks, reds, yellows, greens. No other schedule is needed for non-constraints.
3. Load control and due dates. A new order’s date is quoted from the load already on the constraint: today + load on the constraint in days + half the production buffer. Orders released at the rate the plant finishes them keep the work in process constant.
4. Improvement from the reds. Record why every order went red. Each month, the top reason on the Pareto becomes the improvement project. If more than about 10% of open orders are red, the protection itself is in trouble: look for an emerging bottleneck or a policy.
Ford controlled work in process with space: a line stops when the space between stations is full. Ohno replaced space with containers (kanban), which works when products and demand are stable. TOC controls it with time: not between every pair of work centres, but for the plant as a whole, so that disruptions are absorbed anywhere and work is held only where it is needed, in front of the constraint.
Pause and think: a plant halves its release time and on-time delivery rises from 60% to 88%. Then it stops rising. What does the standard solution say to do next?
- Halve the buffer again, since the first cut worked so well
- Look at why the reds went red, and remove the top reason
- Release earlier again for the orders that were late, so that they have more time
Choking alone rarely takes a plant above about 90%. The rest comes from buffer management and from attacking the top cause of reds. But sometimes the top cause is not a machine or a supplier: it is an assumption of the solution itself that doesn’t hold in that plant.
Nine assumptions behind the standard solution
The standard solution works so often that its assumptions become invisible. Each lesson in this course takes one that fails in real plants:
| The standard solution assumes… | Where it fails | Lesson |
|---|---|---|
| The constraint stays put | The product mix moves it | 2 |
| Orders are of similar size and content | One huge or unusual order | 3 |
| Touch time is under 10% of lead time | Forging, heat treatment, long processes | 4 |
| One order is one item | Orders of many items that must ship together | 5 |
| Feeder shops can work order by order | Feeders with big set-ups | 6 |
| Everything is made to order | Some products must be in stock | 7 |
| Stock and orders don’t compete | They share capacity and material | 8–9 |
| The customer’s due date is real | The customer’s own project slips | 10 |
| Capacity is roughly enough | Peaks, and very expensive constraints | 11–12 |
The pattern in every case is the same, and it is the pattern of the Thinking Processes: an undesirable effect that the standard solution doesn’t remove, a conflict underneath it, one wrong assumption, and an extension to the solution that breaks it, followed by a check for its own negative branches.
Check your understanding
Answer all five. Get four right to complete the lesson.
1. Orders take 50 days through a plant where their touch time is 5 days. This is mainly a sign of…
- A shortage of capacity at most work centres
- Inefficient operators who don’t keep their machines running all shift
- A flow problem: orders spend most of their time waiting
- Unreliable vendors who deliver late
Ninety per cent of the time is waiting, which better flow can remove without new capacity.
2. Why can the release time usually start at half of today’s lead time?
- Touch time is small, so half the lead time is still mostly buffer
- Customers will accept half the lead time once they hear about TOC
- The constraint runs twice as fast after choking
- Half is the legal maximum for work in process
When touch time is under 10%, half the lead time is still mostly protection. Lesson 4 shows what changes when it isn’t.
3. A new order’s due date is quoted as…
- Today plus the standard lead time of the industry
- Whatever date the customer asks for
- Today plus the full production buffer, whatever the load on the constraint
- Today plus the constraint’s load in days plus half the production buffer
The load in front of the constraint decides when the order can pass it; half the buffer covers the way to completion.
4. More than 10% of open orders are red. This signals…
- That the buffer colours are calculated wrongly and should be reset by planning
- A threat to the protection system: an emerging bottleneck or a policy
- That everyone should expedite harder
- Normal variation
A healthy system has few reds; too many means something systemic, not one order’s bad luck.
5. TOC controls work in process differently from Ford and from kanban because it uses…
- Space between every pair of stations
- Containers at each work centre
- Time, for the plant as a whole, with work held where the constraint needs it
- A detailed schedule for every resource, recalculated by the system every night
Time-based control absorbs disruptions anywhere and suits unstable mixes, where containers don’t.