Voorraadsystemen voor lumpy demand.
Report 98.3.LT.5041, Transport Technology, Logistic Engineering.
In this report several methods which can handle products with a lumpy demand
pattern will be described. Lumpy demand is the demand for a product that has an
irregularity in time and in volume; periods of large scale demands will
alternate with periods of low scale demands or even no demand.
Two methods which are not based on a probability distribution for the demand of
products will be described. These methods are to a large extent based on
exponential smoothing, but are in contrast to exponential smoothing suitable for
an intermittent pattern.
The other methods are based on the stuttering-Poisson distribution. The
stuttering-Poisson distribution is formed when the distributions of arrivals and
the volume of the demand are combined into one distribution.
The methods based on the stuttering-Poisson distribution can be divided
into three parts: methods that use different distributions for the ordersizes,
methods that will filter out large scale demands and so result a better
performance, and a method that is based on a single demand approach.
In this report methods arre described only, they will not be compared to
each other or tested on their performance in practice. For this there a
following research project has to de done.
Reports on Logistic Engineering (in Dutch)
, TU Delft