Demand Management by Aggregation: The Case of Spare Parts

Demand management is a hierarchical process encompassing various levels of decision making. It spans from inventory control at the individual order line level to strategic planning. Aggregation offers a distinct opportunity to improve demand management practices, which essentially rely upon a combination of information along three dimensions: products, customers and time. Product and customer aggregation have been extensively addressed in the pertinent academic literature and equally considered by practitioners and software package developers. However, neither the time dimension nor the combined effects of aggregation across the three dimensions have been adequately studied and empirical evidence in this area is lacking. In this paper, we consider empirically the effects of aggregation on demand management. An extensive dataset composed of 18,937 series related to 2,422 stock keeping units (SKUs) is utilised for this purpose and the simulation structure allows the assessment of 27 realistic aggregation scenarios. Our work emphasises the case of spare parts where the intermittent nature of demand renders aggregation a natural strategy to reduce dispersion and increase predictability. The results allow insights to be gained on the effects of aggregation and tangible suggestions are being made to demand planners. The implications of this work for the theory and practice of OM are explicitly addressed and we conclude with an extensive discussion on the next steps of research in this area