Centre for Strategy and Performance

Measuring Complex Systems

Systems encountered in the real world are complex. Complex, in this context, means systems that exhibit variety, connectedness and uncertainty. The uncertainty arises because of the presence of variety and connectedness. In the area of manufacturing and supply chains, the presence of complexity is a barrier to their effective management. Understanding how complexity affects such systems would allow for a focused approach to improving their performance.

There is one type of complexity, called Operational Complexity, that lends itself to being measured. Theory suggests that an appropriate metric is the average rate at which the supply chain or manufacturing process generates information i.e. the entropy of the process. Theory also predicts that operational complexity shows itself through the formation of queues – these can be of either products or information. Thus the presence of complexity turns the operation into a sort of obstacle course. However the presence of queues means that the system is capable of being directly measured, by observing their dynamic behaviour and its causes.

Complexity can be attacked either through simplification or by trying to control it. The former is usually associated with capital expenditure, the latter is a revenue cost. However complexity is neither good nor bad. For example mass customisation involves deliberately expanding the complexity of the product range to offer customers greater variety. The key is to be able to do this without raising prices. Therefore one can differentiate between ‘good’ complexity – complexity the market will pay for and ‘bad’ complexity that merely involves additional cost.

Over the last seven years, measurement exercises have taken place in factories and supply chains. In all cases it has lead to the identification of areas where operational improvements can be made.

Links:

References

Frizelle, G. 1998 The Management of Complexity in Manufacturing. Business Intelligence

Casti, J. 1994 Complexification. Harper Collins New York

Calinescu, A., Efstathiou, J., Scrin, J., and Bermijo, J. 1998 Applying and assessing two methods for measuring complexity in manufacturing. J Operat. Research Soc. 49 723-733

 by Gerry Frizelle, <gdmfeng.cam.ac.uk>