The application of filter theory to the study of supply chain dynamics

Recently, there has been a rebirth of interest in supply chain demand amplification, a trend accelerated by the availability of a wide range of simulation packages. However, although this approach provides some guidance on design improvements possible in a given situation, it rarely offers insight for the future. In the present paper the use of simple filter theory is proposed to help bridge this gap. The example chosen to illustrate the approach is a three-echelon system in which there are factory, distributor, and retailer activities. The results vividly confirm an optimal design previously obtained via a multi-attribute utility technique (MAUT) expert system. However, the knowledge gained via filter theory should improve yet further the effectiveness of the expert system. This is because the sequential steps to be followed when varying the echelon dynamics as part of the search procedure can be greatly improved. The paper concludes by showing how simulation results might be used to confirm the supply chain dynamic design which will minimize stockholdings in the presence of demand fluctuations. However, it should be noted that in common with the successful application of systems dynamics techniques in production-distribution systems generally, the solutions are most applicable to the medium-term operations horizon. The latter term may need re-definition for use in ‘lean’ supply chains. Our intuitive reaction is that a scientific definition may well turn out to be a multiple of the largest remaining process lead-time in the slimmed down supply chain rather than being the customary arbitrary choice of, for instance, a 12-month period.