The Five-Step Process of TOC

The Five-Step Process of TOC

Where Do We Start?

Welcome to our walkthrough of the five focusing steps of TOC. These steps form the foundation from which mostof the body of knowledge has been developed over the past decade. We have added some empiric case studies to illustrate the concept of each step in more detail. These cases are drawn from our involvement with companies around the world, to implement TOC as their guiding management philosophy. Although the examples may not replicate your own environment exactly, they serve to illustrate that TOC is a real, tangible approach with deliverable results – giving better results than any other performance improvement approach you might consider.

Before we can start our fascinating journey into the world of TOC, two things have to be considered carefully.The first issue is the question of objectives. Any system is created to achieve a specific objective. Some systems have for-profit objectives, others might have for-cause objectives. For example, a set of traffic lights at an intersection has an objective of regulating the flow of traffic through the intersection.

Without an objective, the system would behave in a random fashion, with no regulation or direction possible. If we did not understand the objective of the set of traffic lights, we might have put up some large flashing Christmas decorations instead – much nicer to look at but utterly unsuitable to meet the objective.Before we can apply the five focusing steps of TOC, we should have a very clear idea of what the objective is that our system is supposed to achieve. What objective must your department or company meet?Write this down. You will be referring back to this objective as we move through the steps.The second issue is the definition of the system. Controlling and improving the performance of systems is the biggest application of TOC. The fundamental tenet of TOC is that the system is more than the sum of its parts. The system is capable of achieving more than what the individual parts are capable of doing on their own. That is the reason why the system was created in the first instance! We should be clear on the attributes and boundaries of the system to which we are applying TOC. Although the constraint might eventually be found to be outside the system, we need to understand where we are focusing our attention. Jot down the attributes of the system that you are examining. It is useful to do this in the context of the traditional systems perspective:

What are the inputs to the system?

What is the transformation process?

What are the outputs?

How do we control the system’s behavior?

Where’s the Constraint?

Good question! Why assume that we really have a constraint? The proof that constraints exist lies in the inverse of the question. But let us start with the universally accepted definition of a constraint.

A constraint is anything that significantly limits the performance of the system relative to its goal.

If you have been able to identify the boundaries of the system that you are concerned with, as well as the objective of the system, you should ask yourself the following question:

Is this system producing an infinite amount of what the objective prescribes it to do?

For example, if the objective is to be profitable, are you having more and more of it? Or if the objective is to have nothing of something, like the incidence of crime, how close are we to it? If you are not experiencing the expected outcome, then the system is hobbled by a constraint! Once we have clarity on the objectives, system boundaries, and the existence of constraints, the five focusing steps work a lot better!

You might be interested to know that we deal with three different kinds of constraints. The first constraint type is of a physical nature. For example, if I am in charge of a warehousing and distribution center, I might be constrained by the availability of delivery vehicles.The second type is defined as a policy constraint. These are the rules that the organization invented to regulate and guide the behavior of the system. However, with changes in the environment around the system, these policies might become less appropriate to moderate the system’s behavior, and influence the performance of the system negatively.

The third constraint is driven by human behavior. People are comfortable with set ways and predictable outcomes of actions. They will do the things that make sense to them, and which provide them with a sense of security and control over their environment. Even if the environment changes, they will frequently continue to behave in terms of what gives them security and control.There is a link between these constraints. Firstly, when the system is created to meet a specific objective, the system itself generates the measures of its success. (For example, the number of drivers ignoring the red light. Before the lights were installed, no such measure existed for that intersection). The measures dictate the behavior of people involved in the system. Once the measures exist, people modify their behavior to match what the system dictates.

(Stop when the lights are red, go when they are green).

If the measures are in place long enough, the behavior becomes second nature. People display the same behavior irrespective of the continued existence of the system. This is when human behavior becomes a constraint – rote behavior irrespective of the fact that it might now be irrelevant to the success of the system.Keeping the three kinds of constraints in mind, let us discuss the first step.

Step One: Identify the constraint

Identifying the constraint is half the battle won. When talking to the top management team of an organization, we ask them to individually describe the constraint that limits the performance of the organization at this point in time. In virtually all cases, we get as many different answers as there are delegates in the room. The reaction to this is frequently one of stunned surprise, and dramatically highlights the major differences in perspective as to what the true constraint to improved performance is. The answers provide a good basis for discussion to try and identify the constraint.So, how does one identify the constraint? If you suspect a physical constraint exists, it is fairly easy to find. Go look where the work is piling up. In administrative, distribution and manufacturing systems this is a fairly straightforward exercise. Except if human behavior gets in the way!

For example, The TOC Center was called in by an automotive exhaust manufacturing facility. We knew that there was a physical constraint somewhere, but the work in process did not accumulate anywhere. Then we discovered that operators were being measured on utilization rather than output, which caused them to carefully and informally regulate the flow of work between workstations to ensure that no one looked idle. A policy constraint that generated a human behavior constraint!However, physical constraints are relatively simple to identify. It can be done with computerized simulation, direct observation or discussion with people that are close to the action.We were involved with one of the largest gold mines in the world. Once management grasped the basic principles of physical constraint management, they identified a simple physical constraint – one we cannot divulge due to client confidentiality. It had been there for decades! Top management assured us that the removal of the constraint would increase output by 25%! This comes with NO increase in operating expense.

Applying the same TOC principles to other mines have resulted in them becoming the lowest unit cost producers in their class in the world by removing physical constraints. One thing we have learned through the years of applying TOC, is that nothing should be taken for granted when looking for constraints.

Policy constraints are more difficult to find. They are the rules by which the organization functions. Frequently they are the creation of top management, and have an aura of sanctity about them. In many instances, it takes the innocent questions of an outsider like The TOC Center to point out that the emperor indeed has no clothes! Here the TOC Systems Thinking Process - one of the most powerful analysis tools invented to date, is indispensable to identify policy constraints in a logical, understandable manner. It is accomplished by studying the behavior of the system, and deriving the policy constraints from the behavioral symptoms.

The last category, human behavior constraints, can be found by:

Identifying the measurements that the system uses to moderate the behavior of people Measuring the attitudes of people towards the system, its demands, and isolating the dissonance We have found that apparent illogical and irrational behavior makes eminent sense once the measures are clearly understood.Unfortunately, it is difficult to highlight human behavior constraints if you yourself are part of the system. The TOC Systems Thinking Process is indispensable to clearly identify the human behavior constraints.

Step Two: Exploit the constraint

Now that you have found the constraint, what will you do with it? The obvious answer is to get rid of the darn thing! This is true for policy and human behavior constraints. We would like to resolve those as fast as we can. We don’t want to be saddled with an inappropriate policy, or a debilitating human behavior constraint. However, physical constraints are somewhat different. Our experience has taught us that many cases where focusing on the physical constraint, and making sure that it performs to the maximum of its capability at all times, generated 20% to 100% of additional output with little or no increase in operating expense.

Exploitation is accomplished by re-assessing the way the constraint is utilized: the way that it is applied to make the right products and how it is supplied with work at all times. We have redesigned jigs and fixtures, reconfigured scheduling systems, changed materials handling systems, rescheduled shift patterns, re-educated employees, and modified machinery to ensure that the constraint is fully exploited.

However, to sustain such exploitation, it is frequently necessary to change the measures and policies used to regulate the behavior of the system. This sounds a lot like policy and human behavior constraints!

To summarize:

Make sure that the physical constraint always has something to do

Make sure it is doing whatever it is supposed to do as fast as possible

Remove all sources of delay or decreased output

Ensure that the measures around the constraint support the behavior you expect people to exhibit

Step Three: Subordinate everything else to the constraint

Our experience has proven that this is the most underrated of the five steps, and the most difficult to accomplish. To ensure that the constraint works well, it is imperative that the rest of the system is geared to support the performance of the constraint. This implies that the performance of each individual part of the system should be subordinated to the constraint.

The activities, scope and speed of every other part of the system should be geared to the requirements of the constraint. Although this sounds quite simple, the implementation of this step is frequently fraught with many unexpected difficulties. Individual measures, localized departmental objectives, inter-departmental rivalry and other issues make this a fairly difficult outcome to achieve. Some pointers might be of help here:

Make sure that everyone understands and agrees with the validity of the concept of supporting the

constraint Communicate the measures of success for the constraint and the non-constraints to everyone Make sure that the non-constraints understand how they are to support the activities of the constraints. For example, balancing the flow and not the capacity, removing conflicting measures, and linking gating operations to the constraints go a long way to achieving this In particular, attention should be given to the possible occurrence of peaks in demand on non-constraint resources, and how that will affect the constraint More detail can be found in the “The Goal” by Eli Goldratt and Jeff Cox, and “The Race” by Bob Fox and Eli Goldratt. Both books can be ordered from The TOC Center.

Step Four: Elevate the constraint

Only, and only when the entire system has been geared to support the activities of the constraint, and the maximum output has been achieved, should consideration be given to elevating the constraint. This implies increasing the capacity of whatever the constraint may be – more production capacity, more distribution capacity, more market etc.Our experience indicates that organizations can typically gain 25 – 100% of additional output without any notable increase in operating expenditure or investment. Elevating the constraint is frequently the easiest of the five steps to execute. Once the constraint is elevated, the system’s performance increases. However, when this occurs, the constraint moves to another point in the system.

Step Five: Go back to Step One

Once the constraint has been elevated, it is fairly easy to sit back and enjoy the performance improvement that has come about. Since TOC is a process of continuous improvement, one has to guard against complacency derailing the process. It has to start again from Step One. A concerted effort is needed to identify the new constraint, go exploit it and subordinate the rest of the organization in a different configuration.Our experience indicates that the positive results achieved in the first rounds of the five-step process provide a lot of momentum to sustain subsequent rounds of constraint elimination.

However, many companies who implement the five-step process find that constraints tend to move from one place to another in an unpredictable way. This makes for unstable planning. To alleviate this phenomenon, we developed a new approach, called the Throughput Operating Strategy to provide rapid, predictable increase in Throughput without sacrificing long-term stability.