Operational expense is all the money the system spends in order to turn inventory into throughput.

These definitions help emphasize the waste involved in making product for finished goods inventory. Not only have you sacrificed today's dollar for a promised dollar in the future, but, more importantly, investment in the finished goods inventory will very often stop you from developing and releasing new products – which is a death knell in the current business world.

Bottlenecks and constraints

In most businesses, the key to making more money is to increase throughput (while simultaneously reducing inventory and operational expenses). What stops throughput? Bottlenecks or constraints.

Most processes have constraints. If the constraint is a physical bottleneck, you might be able to find it by looking for a pile of work in front of it. Goldratt points out that constraints (or bottlenecks) are not good or bad. Bottlenecks can be very useful in process improvement work, but they do need careful management.

An hour lost at a bottleneck is an hour lost to the whole plant. That is

You lose time at bottlenecks (among other things) by having the bottleneck work on rubbish. If you can stop the bottlenecks from processing rubbish, you can save that time for the whole system. Rule: put inspection in front of bottlenecks.

He makes several important points.

• The Theory of Constraints gives us a new set of rules for how to identify which problem to work on and how to go about it.
• Always put your quality control in front of your bottlenecks. There is no point having your bottleneck work on defects.
• The majority of constraints are not physical constraints. The vast majority of constraints that prevent us from making more money are our own policies – the way we do things – our own standard operating procedures. Surely, these should be the easiest things to change. As we have seen throughout Principle 4 – the barriers we build in hanging on to the way we do things and demanding people work harder, are significant obstacles.

Service companies should not reject this by thinking it applies only to manufacturing. Many service companies set their processes up like factories; eg banks, law firms, insurance companies.

Service company example

1. You identify your bottleneck as your leading technical expert (eg, a lawyer of a person building your database or web site)
2. You make certain you get every hour you can out of them – no meetings or training (or meetings and training during lunch or weekends or periods of slack) Beware! Too much of this breaks Principle 7 'Enthusiastic People' and will lead to burnout. The steps below are better.
3. Everything else will revolve around the person. Put you quality control in front of the person. Have someone else check what goes to the person her to make certain it is accurate and valid. Do not have them work on tenders (you might not get them) or freebies.
4. Add capacity. Add helpers (who are cheaper and lack to expert's skill) to do the legwork or rough work to which the expert will add the finishing touches. Eventually add another expert.
5. Check to see if the person is still a bottleneck.

Queues and waiting

Working out where thing spend their time can greatly assist you to determine what part of your process you should work on to get most benefit.

Consider a file in a law firm. The percentage of the total cycle time that the file is actually being worked on is usually very small. Most of the total time the firm has the file, it is not being worked on. It is either waiting in a queue with other files for someone to work on it; or it is waiting for information.

Most companies make the mistake of trying to reduce cycle time by reducing the time a file (or part, or inquiry) is being worked on. Wrong! Tackling the queue or the waiting time gives much bigger gains. For example, if one of your bottlenecks is waiting for information, move it to be as early as possible in your process. Ie, request the information as soon as you possibly can.

Inertia

The inertia of tradition in your own policies and standard operating procedures are often your only bottlenecks preventing you from getting more throughput and making more money.

Every time you do something for the first time, you establish a tradition. Because it is tradition, it has become "the way we have always done it". Even if you or no one else remembers why.

There is an anecdote about ceremonial guns – the ones used in military parades. The story is that the gun crew (of seven) roared up in their truck with the gun mounted behind on a trailer. They leapt into action, uncoupled the gun, swung it around and fired off their twenty-shot salute. One man stood at attention behind the action the entire time. A dignitary pointed out the immobile man and asked, "what does that person do". A flurry of questioning revealed that no one knew. A month later, the dignitary received her answer in a fax. "He holds the horses." It was 1999. Horses had not been used with artillery for almost 80 years. Inertia had kept the process unchanged.

You too have processes where people are doing things that are no longer required. Every time you change a process, you make redundant many actions of the old process. You must clear up such redundant actions. They can be your most significant barriers to reaching your Goals and objectives.

Purpose

Another barrier to having your improvements make a real difference is that it is often difficult to know the purpose of something when you only look at each component. Consider a car. Its purpose is transport. Could you work that out by looking at each part?

People often disassemble systems in order to identify their parts. In doing so, they may lose site of the whole or the purpose of the whole. A system is an interacting whole consisting of independent parts. You cannot understand the whole by understanding each part. You cannot improve the whole by improving each part. The net output of the whole — for better or for worse is the result of the interactions. To understand the whole you must understand the interactions. You cannot understand the whole without understanding its purpose. This means that `improvement' that is not system based is not improvement and that 95 percent of change is not improvement.

Policies are usually the main constraints

Usually company policies are the main barriers to achieving the Goal.

For example, the local fast food restaurant El Turk, is in the food court at the shopping mall. El Turk opens from 8:30 am to 5:30 pm. A long 9 hour day for them. A large crowd occupies the food court from 6 pm to 11 pm for its evening meal. Other fast food stores do a huge trade. But not El Turk, it is closed. Why? A policy.