Hello,
I am trying to build an (r,Q) inventory control model , to be implemented in a local manufacturing firm. In my model, r is identified by studying the demand trends and fitting normal distributions, while Q is identified by genetic algorithm minimising an inventory cost function.
Firstly the list of parts for re-order are first ideintified, after which a genetic algorithm co-optimises the batch sizes across all parts on re-order, considering factory capacity and demand constraints.
In my simulations, I find that the Q values suggested by my algorithm are realistic and in-line with what is required, however, it turns out that the model identifies all the parts which are less likely to be ordered by the production team.
Would anyone be able to shed any light on how to control this ?
Dear Akhilnandh,
an (r,Q) model is a reorder point model, where r is the reorder point. When the inventory position of an item reaches r, the Q units are ordered. Thus, I don't understand what you mean when you say "Firstly the list of parts for re-order are first ideintified". In a (r,Q) policy the part to be re-ordered is the one that reaches the reorder point.
- Badges
- Science topic
- Similar topics
- Mathematics