Cynax
Menu
https://www.cynax.nl/en/cases/demand-pulse-cases
Cynax Simplifying Information Technology

    Van Schappen

    Van Schappen, a tool manufacturer had been receiving many complaints. Orders are rarely delivered at the agreed time,and the delay varies from a few weeks to even several months.

    In addition to these complaints, Van Schappen had itself also identified a number of problems: inventory level was often too low, extra costs because of urgent orders, and, on the 2010 balance sheet, an undesirable high value of remaining inventory. Van Schappen decided to use Demand Pulse (DP) to deal with these problem areas.

    problem definition

    During the implementation of Demand Pulse it quickly became clear that Van Schappen’s estimates of the demand for their products were inaccurate.It turned out that their
    method worked reasonably accurately only for the first few weeks, with an error percentage of 6%. But the longer the forecasting period, the greater the margin
    of error: no less than 24% at the end of the period! The sales pattern
    of a specific Stock Keeping Unit (SKU) can vary widely, as shown by the graph.This shows is that, for example, for 95% of the time, between 0 and 45,000 units will be sold on a
    particular date. The smaller the range of the number of units, the less certain the forecast.

    Our delivery times have improved dramatically, even with lower inventory levels.
    - H. van Schappen -

    resultaten

    In reality, in 2010, ‘P1’ was subject to many inventory shortages or stockouts (where the red graph falls below the zero line in the diagram above) that – due
    to the great pressure on the production department - ultimately led to an enormous inventory level at the end of 2010.
    If Van Schappen had calculated the parameters using Demand Pulse in 2010, then halfway through the year a stabilisation point would have reached with hardly any stockouts and far fewer inventory fluctuations. Even with less Inventory…

    Using Demand Pulse:

    • There would have been an average inventory of 464,000 units over the whole of 2010, with a standard deviation of 145, 000 units.
    • This inventory would have stabilised in the second half of 2010 with an average of 211, 000 units and a standard deviation of 88, 000 units.

    In reality:

    • There was an average inventory of 650, 000 units over the whole of 2010, with a standard deviation of 186,000 units.
    • In the second half of 2010 there was an average of 391, 000 units  with a standard deviation of 186,807 units.