Problem Structuring

Most real world problems are complex and ill-defined. Before they can be subjected to analysis they need an element of structuring. Some problems can be well defined, but these tend to be the less interesting ones. Thus, if you have an interesting problem, it is usually poorly understood and needs to be structured before it can be analysed.

Often problem structuring can be the most useful single phase of any study from the point of view of the client. This is useful to those looking to economise, since it is the phase of the project that necessarily happens at the beginning.

Delineating the problem involves answering two questions:

  • The first question concerns the scope of the problem - what things do I want within the problem boundary and what things do I want outside. This is often not an easy decision and setting the boundary of the problem often leads to the realisation that important aspects of the problem have been placed outside that boundary and need to be brought back in.
  • The second question concerns the elements of the problem and the relationship between them. This is the more difficult part of the problem. There is likely to be some kind of link between almost all of the elements of the problem, even if it is only that they are part of the problem. An important skill for the analyst is being able to decide at what point to decide that the relationship between two elements of the problem is sufficiently tenuous that it can safely be ignored.
At Larrainzar Consulting we use a number of different problem structuring techniques to help our clients. These are not the only techniques available, just those which we find the most useful.
Stacks Image 132
Houston, we have a problem….
Jim Lovell, Commander Apollo 13

Mind Mapping

A simple example of a mind map is shown to the right, with a more complex example below. Mind Maps were originally developed by Tony Buzan and are often used for learning. We use them as one of the most rapid and helpful methods of quickly establishing the scope of a study or problem and agreeing it with the client.

The nature of the mind map tends to spark associative thinking, which is ideal when establishing scope. The format is also simple, requires little explanation and is easily understood whether the client prefers words or images. Thus the client can get an early overview of the elements of the problem that will be included in the study.
Stacks Image 266
Stacks Image 264

Influence Diagrams

Influence diagrams are a way of showing how the different elements of systems inter-relate. They are a tool used in the early stages of building Systems Dynamics models. The influence diagram shows the direction of cause and effect in the system and can show where positive and negative feedback loops are likely to occur. On the diagram on right the causal link is shown as positive when an increase in the variable at the base of the arrow tends to increase the variable at the head of the arrow. A negative link shows the opposite. Feedback loops are shown by the small circular arrows. A 'B' loop is a balancing loop and tends to restore the system to its original state. An 'R' loop is a reinforcing loop and tends to accelerate changes from the original state.

Often the problem with a complex system comes from not understanding that pushing hard on one element of the problem may give rise to unforeseen feedback effects that produce changes in the opposite direction to that intended.

As an example, when a project is running behind managers often react by moving new resources onto the project. This can give rise to negative effects, since the new entrants to the project have to be trained by those already working on it, taking time away from their main task, whilst the new entrants are also likely to have higher error rates, thus forcing the team to spend more time correcting errors. Thus more people on a project may lead to it running further behind schedule. This kind of behaviour can easily lead to a positive feedback loop, where managers react to the ever-slower progress by adding ever-more resources to the project.

Influence diagrams are a way of understanding how positive and negative feedback loops influence the behaviour of the system of interest. They can give important insights into the reasons why the system behaves as it does. These kind of feedback loops are important to a huge variety of systems.
Stacks Image 240
Stacks Image 212

Soft Systems Methodology

Soft Systems Methodology or SSM is a method developed by Peter Checkland over the past 40 years or so. It concentrates upon understanding human activity systems.

Often in systems thinking it is assumed that human systems are similar to systems in biology or ecology. Thus, there are inter-related parts, the interactions have important implications for the behaviour of the system as a whole and there is emergent behaviour.

There is however an important difference between human activity systems and other kinds of system. Human systems have a purpose, or more precisely a number of purposes. Understanding what those purposes are and how views of the purpose of the system differ between different groups is often fundamental to understanding how to improve the system.

For example, most organisations have a pay system. From the point of view of the workers the purpose of the system may be to reward good performance and ensure fair pay. From the point of view of the human resources department the purpose of the pay system may be to avoid litigation and retain staff. From the point of view of the management the purpose of the pay system may very well be to control costs. One system, at least three different purposes.

To help understand the different purposes of a system, Checkland developed what he called root definitions of the system. Generating root definitions makes use of the CATWOE mnemonic. These are the things that need to be identified as part of the root definition of the system.

Root definitions of systems need to be built with those who are involved in the system. Much of the power of SSM comes from its ability to focus and clarify the thinking of those with the ability to do something about the system. If the clarity exists only in the mind of the analyst, it is of no use.

Rich Pictures

Rich pictures are also an important part of SSM, and can be used as a problem structuring tool in their own right. A rich picture shows the activities carried out as part of the system and shows the inter-relationships between them. Rich pictures, like root definitions need to be developed in conjunction with those involved in the system. Unlike the CATWOE-based root definitions, the rich picture concentrates upon more concrete elements of the system, whatever they might be.
Stacks Image 182
Stacks Image 134