Inventory and Demand Analysis

ABC Analysis can be used to assign the appropriate level of control and review frequency based on the annual dollar volume of each item. Classical ABC Inventory Analysis places:

  • greater expenditure on supplier development for A items than for B or C items
  • tighter physical control on A items than on B and C; cycle counting A items more frequently than C
  • greater expenditure on forecasting A items than on B or C
  • different replenishment or order policies for A items than on B or C

 C items are often handled with simple techniques of min/max or reorder point. Some practitioners make the mistake of trying to apply kanban to either A or C items. What is missing is an understanding of demand linearity (or demand variability). ABC Analysis is typically based strictly on volume, or annual value. This approach would then treat both very predicable and highly volatile A items in the same manner. But one size doesn’t fit all… What’s missing is a little statistical understanding of the item demand pattern. Does consumption happen smoothly and regularly or are there big spikes in demand? When you take the standard deviation of the demand history and plot it against volume you get a demand segmentation like so …

 

 

 

 

 

4 thoughts on “Inventory and Demand Analysis

  1. Hello

    I am trying to do an ABC analysis of a product portfolio based on volatility of demand.
    Wich levels could delimited A, B and C?? because usig pareto 80% for A is very high…

    What could be an acceptable volatility for product?

    Thanks a lot

    Bests regards

    jean-philippe

  2. Michael,

    You’re right, ABC is a start. Considering demand volatility attempts to get at stock out risk. Another dimension is certainly lead time – how quickly can you recover from “oh shit we’re out”.

    In a recent publication on the Supply Chain Digest web site there is the 2006 WMS RFP which contains a short list of cycle count requirements.

    Cycle count
    1 Generate counts through errors
    2 Generate counts by A/B/C
    3 Download counts from host
    4 Create counts by ranges of items, locations and price
    4.1 Opportunistic cycle count during picking

    Other examples of opportunistic counts include:

    1 Putaway – while you are there count the bin above, below, left or right
    2 Zero bin – systems says the bin is empty, is it? This is a very quick count.
    3 Negative stock

    Regards,

    Lawrence

  3. Can we expand a bit here? What are the “accepted” methods of cycle counting/analysis? I have a comptroller who has insisted a locally purchased lock washer, valued at $0.003 each with relatively low consumption is an “A” item. She has gone from complaining that we are behind in cycle counts to we count too much and waste resources. Her last response to my questions of how she arrived at her classifications is “I’m the Comptroller, blah blah blah.”

    As a past Materials Manager with education under Ollie White and APICS, I learned the standard Pareto analysis is not always the best. It is a starting point only. Balancing this with risk and item type is best. Something you can buy in town offers less risk to the business than something purchased over seas. Therefore, something that would classically show up as a “c” might be elevated to a b or a under a situation where lots of inventory is a bad thing.

    Anyway, do you know of widely accepted methods of ABC analysis and cycle counting that is more than the simple “count the items on the daily cycle count list”? I’m looking for descriptions of activity based cycle counting, or methods where lower than c classes are used for commodities such as fasteners, etc..

    Any modern info would help.

Leave a Comment ...