A chain is only as string as its weakest link.
A bit of a cliché, but the trouble is we cant always wave the weakest link goodbye, we have to manage it.
This is where the Theory of Constraints (TOC) comes in (constraints in the sense of problems bunging up a process, not a spatial designation). It was introduced in the 1984 book The Goal by Dr. Eliyahu M. Goldratt. It was based on an earlier idea called the theory of bottlenecks but Goldratt greatly popularised it in the English Speaking World. It really is one of the very few books id recommend that any manager has to read.
The book is unusually written as a novel, of a manager struggling to save a failing factory. Through a series of dialogues the protaganist works out the solutions himself through challenging conventional wisdom. He works out that the business really has only one goal, not to maximise output but to make money. He then set out to work out how to measure how the money he was spending on costs translated to sales.
He figured out he needed to measure three things (ill use the modern treatment rather than the early version in the book):
- Throughput is the rate at which the system generates ‘goal units’ e.g. money through sales.
- Investment -the money tied up in the system. This is money associated with inventory, machinery, buildings, and other assets and liabilities.
- Operational expense the money the system spends in generating “goal units.” – translating investment to throughput.
The same approach can be used in public services if the concept of sales is replaced by one of the ‘goal output’ of the service.
At the heart of this is a critique of traditional concepts of cost accounting. If a unit needs to save money cut costs. The problem is that across the board cuts can just as easily lose you money – or of course make the service much less value for money.
When cost accounting was developed in the 1890s, labour was the largest fraction of product cost, and hours if work could be highly variable. Cost accountants, therefore, concentrated on how efficiently managers used labour since it was their most important variable resource, & many managers are still evaluated on their labour efficiencies, and many “downsizing,” “rightsizing,” and other labor reduction campaigns are based on them.
Now, however, workers who come to work on Monday morning almost always work 35-40 hours or so; their cost is more fixed rather than variable.
This approach is important because it allows focus on how of constrained resources restrict the achievement of an organisations goals.
Imagines a body has to process 1000s of forms – a not unfamiliar example to planners – before it can progress a plan.
Now imagine there was only one workstation and only one person to enter them – in the language of the Theory of Constraints the both the number of workstations and the number of operators would be bottlenecks – constraints.
Now lets say that it was found that to process the forms within 1 month you needed an extra 5 workstations and 5 operators. Imagine the operators were hired on a temporary contract. The constraint would not shift to the workstations and no longer the operators. Lets now say that 6 workstations were leased but sat in a box because there was noone to set them up.
Along comes a cost accountant keen to downsize. They send the workstation back, they are overhead. They sack the operators. Now by their books they have radically improved efficiently. However the one original person will now have to take 6 months to enter all the data. During the extra 5 months all of the co-workers salaries have to be paid, and the output is 5 months late. The real costs, measured in terms of achieving the goal, costs have not gone down they have dramatically gone up. Those staff and the workstations was not an operational expense but an investment.
This is one of the reasons why organisations should operate on the principle that everything is a project and of zero-based budgeting. Otherwise overhead costs of staff are not properly accounted.
The Theory of Constraints is based on the premise that the rate of goal achievement is limited by at least one constraining process. Only by increasing flow through the constraint can overall throughput be increased.
Assuming the goal of the organization has been articulated the steps are:
- Identify the constraint (the resource or policy that prevents the organization from obtaining more of the goal)
- Decide how to exploit the constraint (get the most capacity out of the constrained process)
- Subordinate all other processes to above decision (align the whole system or organization to support the decision made above)
- Elevate the constraint (make other major changes needed to break the constraint)
- If, as a result of these steps, the constraint has moved, return to Step 1.
The concept of subordination means that once a bottleneck has been identified improving throughout anywhere else in the business process wont increase the output overall one iota. Things will still stack up infront of the bottleneck.
Elevating the constraint means that if you have done all of the above and it is still a constraint you need to invest to increase capacity, but only to the extent that it remains the constraint otherwise the investment is waisted.
Using these principles you will find that constraints will often switch and switch back, you need to be constantly improving.
A major way to avoid constraints being a problem is to utilise the concepts of ‘pull’ management introduced in the previous lecture. This is through the idea of ‘Drum Buffer Rope‘.
The rope is the ‘pull’ of resources from upstream. The drum is the rate of the business process. The buffer is what protects the drum from being held up by the constraint.
Think of poor dad in this diagram.
To keep going he cant go faster than the pull of the rope. Unless he slows down to a crawl the rate he goes out will be determined by the variability of how quickly his toddler behind goes, although the impact of this will be significantly reduced if the toddler is at the end of an elastic cord, in which case his speed will be defined by the length of the cord.
The idea of a buffer is to protect a process from a variable constraint.
It is surprising perhaps that this concept has not been used more in planning, especially in infrastructure planning, as infrastructure is a classic constraint.
Heres an example of where $85 million was spent on removing a bottleneck, to no effect on the level of service.
One good example of where it has been used successfully is concerning the upgrading of the East Coast Line. Conventional wisdom held that the main constraint was the notoriously narrow Welwyn Viaduct. This study found it wasnt, and that upgradings elsewhere on the network were needed.
These ideas are just as powerful when looking at constraints in time – project plans – the next section.