demand management - letter: tony wild's march correction article banner

Operations Management Articles Home
Quick Article Search
Contact Operations Management Articles

Operations Management Article Categories:

  • Advanced Planning and Scheduling Articles
  • Analytical techniques Articles
  • Assemble to order Articles
  • ATO Articles
  • Bill of materials Articles
  • Breakeven analysis Articles
  • Change management Articles
  • Computer systems Articles
  • Conferences and exhibitions Articles
  • Continuous improvement Articles
  • Costing Articles
  • Customer relations management Articles
  • Customer service Articles
  • Data handling Articles
  • Demand management Articles
  • Distribution Articles
  • Ecommerce Articles
  • Efficient consumer response Articles
  • Enterprise resource planning Articles
  • Environment Articles
  • ERP Articles
  • Forecasting Articles
  • Globalisation Articles
  • Government and politics Articles
  • Health Articles
  • Human resources Articles
  • Industry sectors Articles
  • Innovation Articles
  • International Articles
  • Internet Articles
  • Inventory management Articles
  • Kanban Articles
  • Lean manufacturing Articles
  • Lean operations Articles
  • Legal and regulatory Articles
  • Logistics Articles
  • Maintenance Articles
  • Manufacturing control Articles
  • Manufacturing resource planning Articles
  • Marketing Articles
  • Material requirements planning Articles
  • Metrics Articles
  • Modelling Articles
  • Outsourcing Articles
  • Performance Prism Articles
  • Planning and scheduling Articles
  • Process control Articles
  • Product control Articles
  • Publications Articles
  • Purchasing Articles
  • Quality management Articles
  • Research Articles
  • Shop floor control Articles
  • Strategic Articles
  • Supplier management Articles
  • Supply chain Articles
  • Systems implementation Articles
  • Training Articles
  • Virtual enterprise Articles
  • XML Articles
  • Y2K Articles
  • Downloadable Products
  •  

    Letter: Tony Wild's March Correction - Demand management Article

    The article entitled: "Letter: Tony Wild's March Correction" is in the Demand management Articles section of Operations Management Papers area...

    ARTICLE DESCRIPTION: Tony Wild's correction in the July/August issue of 'Control' to his letter in the March issue does not go quite far enough. His calculation of Mean Absolute Deviation (MAD) is too large by a factor of 6 as he has inadvertently quoted the Sum of the Absolute Deviations (SAD) - that none of us spotted!) Nonetheless, his letter raised two issues, which may usefully be expanded :

    MORE INFO: Letter: Tony Wild's March Correction Tony Wild's correction in the July/August issue of 'Control' to his letter in the March issue does not go quite far enough. His calculation of Mean Absolute Deviation (MAD) is too large by a factor of 6 as he has inadvertently quoted the Sum of the Absolute Deviations (SAD) - that none of us spotted!) Nonetheless, his letter raised two issues, which may usefully be expanded : The first is that for determination of service-driven safety stocks he would be better to use the standard deviation. While it is true that for a normal distribution the standard deviation can be approximated by 1.25 times the MAD. We should bear in mind that (i) as this relationship is only an average, it is subject to sampling variability, (ii) it only applies to the normal distribution and (iii) it rose to popularity in inventory control only because of the ease which it would be calculated in the early days of computing. Inventory control relies a great deal on averages and the assumption of the normal distribution, of course, but with modern computing power we should use the better measure of variability. One noted consultant in this field pointed out many years ago that the use of standard deviation rather than MAD could reduce safety stocks by about 10%, in his experience. The exponentially smoothed MAD is still used as the denominator of Trigg's tracking signal, but Brown's formula for calculating safety stocks specified a standard deviation. The second is that for inventory control purposes it is much better to analyse the demand pattern for each individual stocked item than to attempt a top-down analysis. I agree with Tony that to try to apportion out an individual demand pattern from that of the group would be doomed to failure. It is difficult to view the example he quotes as anything more than the result of random variability: Sometimes the variance at top level will be greater than the sum of the individual variances, sometimes less. The patterns of demand at individual item level could differ from that of the total group in matters of trend, seasonality and even the possibility that some individual items could exhibit intermittent demand while a more continuous pattern is observed at top level. With modern computing power there is no need to settle for anything less than full analysis by SKU, since that is where the stock-holding decisions have to be made. The classification of demand is an important issue in inventory control because the forecasting techniques, the exception reporting and the method of replenishment are all driven by it. It would be nice to work with stable well-behaved demand data, which conform, neatly to the normal distribution, but unfortunately commercial data can sometimes be astonishingly intractable. Modern forecasting and inventory decision support software, such as the Demand Management system I am currently developing at STG, takes these matters fully into account: It has to be able to deal effectively with the vagaries of real demand data in all industrial sectors. I suspect that Tony's problem arises from a lack of data at item level with which to perform the most appropriate analysis. One can only hope that as the process of data capture and retrieval continues to improve, this problem will soon be a thing of the past Yours sincerely Mike Thomson Forecasting Production Manager - STG

    PUBLISHER: Institute of Operations Management

    Page number: 7 Words: approx. 350

    Vol 25 - No 07 - September 1999

     

    The article can be downloaded in full from the publisher's site i.e. the Institute of Operations Management.

    Thank you for searching on Operations Management Articles.co.uk for Demand management article entitled: Letter: Tony Wild's March Correction in the Operations Management / Demand management Articles and Papers Category.

     

     

    Business Performance Improvement

    © 2006-8 Operations Management Articles