Theory of Constraints: Find and Fix Your Bottleneck

Every process has one step that is slower than all the others. That step, the constraint or bottleneck, sets the pace for the whole process. Work cannot leave the process faster than it gets through the slowest step, however quick the others are.

The theory of constraints (TOC) was set out by Eliyahu Goldratt in his 1984 business novel The Goal. Its central point is that improving anything other than the constraint does not increase output. An hour saved at a step with spare capacity is, in Goldratt's phrase, a mirage.

Why this matters for a small business

Most small businesses respond to being busy by working harder everywhere, or by hiring generally. TOC says to find the one step that limits output and put your effort there. Typical constraints in small firms are:

• One skilled person everyone waits for: the only qualified gas engineer, the owner who signs off every quote, the one designer.
• One machine: a spray booth, a kiln, a CNC router.
• A stage outside your control: survey visits, building control sign-off, a supplier's lead time.
• Sales: if every step has spare capacity, the constraint is in the market, not the process.

The five focusing steps

1. Identify the constraint. Look for where work piles up. The step with the biggest queue in front of it is usually the bottleneck.
2. Exploit it. Get the most from the constraint as it is. Make sure it never sits idle: cover its breaks, keep work queued in front of it, and check quality before work reaches it, so it never processes something that will be scrapped.
3. Subordinate everything else. Pace the other steps to the constraint. Steps upstream that run faster only build piles of work in progress.
4. Elevate the constraint. Only once it is fully used, add capacity: a second person, an extra shift, another machine.
5. Repeat. When the constraint moves, start again at the new one. Do not let old rules become the new constraint.

Worked example

A small fabrication shop makes steel gates. Each gate goes through five steps. Their capacity per hour, measured over a week, is:

Welding is the constraint. On a seven-hour day the shop can make at most 18 × 7 = 126 gates, against demand of 140. Hiring another packer, or buying a faster saw, would not add a single gate.

Each gate earns £35 of profit. One extra hour of welding, from covering lunch breaks for example, produces 18 more gates and £630 of profit, but only 14 are needed to meet demand, so it is worth 14 × £35 = £490 a day. An extra hour anywhere else is worth nothing until welding is fixed.

Find your bottleneck

Enter each step and its capacity per hour. If you are not sure of the figures, count the output of each step over a few days.

Exploiting the constraint before spending money

Step 2 is where most of the cheap gains are. Ways to get more from a constraint without investment:

• Stagger breaks so the constraint keeps running.
• Move inspection before the constraint, so it never works on faulty parts.
• Take non-essential tasks away from the constrained person: paperwork, fetching materials, answering the phone.
• Shorten changeovers at the constraint; the SMED guide shows how.
• Plan its work a day ahead, so it never waits for a decision.