Inventory control in environments with short lead times

Methods currently used for inventory control in industry are with few exceptions developed more than forty years ago. This is for instance the case with methods like reorder point sys-tems, cover-time planning and material requirements planning (MRP). Lead times at that time were typically several weeks and months and order quantities covered typically de-mand over several months. Since then both lead times and order quantities have been re-duced in most companies. Today lead times represent weeks or even a few days rather than months and order quantities cover a few weeks of demand rather than several months’ de-mand.
For various reasons shorter lead times and smaller order quantities make some of the as-sumptions in the models used for inventory control less acceptable and make the models to a lesser extent reflect the planning environment in which they are used. This is for example the case for the assumption that issued quantities are unit-sized and that the demand during lead time can be represented by a normal distribution. As a consequence, the models are expected to perform worse than anticipated and be less appropriate to use
The research presented in this paper concerns the development of enhancements and modi-fications to currently used models for inventory control, making them more suitable and effective in environments characterized by short lead times and small lot sizes. The per-formance of the developed model, as well as those of traditionally used models has been assessed by simulation. These simulations show that inventory control methods used in industry fail to secure the desired service levels that the methods are designed to attain. The simulations also show that by using the developed model the differences between desired and attained service levels can be reduced to fall within limits acceptable in practice.