How to choose the appropriate service level has been described by many authors and logistics professionals. The common basis for differentiating the policy parameters between products used by many professionals is a traditional ABC-analysis based on the sales volume in SEK, i.e. the demand times the price per unit. Even if this basis is suitable for several issues in inventory management it is not the best for differentiating the service level. This since the value of having a high service level for a high-price product is comparatively low whereas the value of having it for a high demand product is high. There hence exist arguments both for a high and a low service level for A products.

The purpose of this work is to 1) estimate the potential savings from differentiating the service level 2) contribute to the understanding of which factors should be used as the base for differentiating the service level 3) develop an easy to use method for grouping the products and determining the appropriate service level for each group. In the paper we suggest a simple method for differentiating the service level using a combined ABC/XYZ analysis based on the sales volume in SEK (as the traditional method) and the demand in units. The economical benefits of the suggested method have been evaluated using numerical examples based on real life demand data. This study indicates that substantial cost benefits can be achieved compared to a system with no differentiation of the service levels.