Decision Making / 4 / 4.9 / 4.9.6

4.9.6 Tame Vs. Wicked Problems—Explanatory Characteristics

A tame problem may be stated as a ‘gap’ between what ‘is’ and what ‘ought’ to be and there is a clear explanation for every gap. Wicked problems have many possible explanations for the same discrepancy. Depending on which explanation one chooses, the solution takes a different form.

As explained above, conventional problem solving methods are inappropriate for addressing wicked problems.

In the process of simulating wicked problems, even if it is limited to mental simulation, it is possible that different analysts will build quite different models. Each model may produce the same, or very similar, patterns of dynamic behaviour. Unless these different models are comprehensively evaluated, it is quite possible they will be accepted as perfectly reasonable depictions of the problem space and used for strategy or policy development, with potential to produce quite different results.

In relation to the scientific rigour required in the development of computational models, and sometimes these will be essential, Homer (1996) makes the following observation:

‘… adherence to the scientific method is essential for developing models that will stand the test of time. Both client and modeler may be tempted at times to abandon the process, especially when the needed evidence [for comprehensive evaluation] is not readily available… the modeler must… not be satisfied with surface appearances… deep evaluation is… the modeler’s responsibility… evaluation-based insights often help resolve controversies about what is really important in a system and what is not (Homer, 1996: 16-17).’

Though we are dealing with qualitative modelling, the philosophy is the same. Identification of what is really important, or highly influential, in a system will drive selection of pressure points or leverage points to which we might apply resources and management effort. If this process is flawed, strategies and policies we might derive from the models are also likely to be flawed.

Because wicked problems have many possible explanations for the same discrepancy, we need to be careful in:

• model building;
• testing; and
• deciding on which explanations regarding:
• system influence, and
• the model behaviour we choose.

These risks may be minimised by the rigorous application of the scientific method:

‘[Real] science is, quite simply, the scientific method, the relentless iteration of induction and deduction, of precise hypothesis formulation and careful experimentation (Meadows and Robinson, 1985: 419).’