Meeting with Stefan 20.04.21 (PC)

All their products are Make-to-Order:

• They hold almost no inventory on Finished Goods (DOI = 12 days) -> they are not expecting us to reduce stock there

• 74% of their inventory value is in Raw Material and Work-in-Progress (DOI = 67 days) -> this is where they want to reduce

Benchmark of their current Forecast Accuracy for the whole portfolio (see screenshot attached)

• Short term (1-4 weeks) = ~85%

• Medium term (4-11 weeks) = ~60%

• Long term (12+ weeks) = ~45%

Stefan is looking into medium and long term forecasts (3 and 6 months) because this is the timeframe when they have to order their raw material (LT = 12-16 weeks).Make-to-Order explains why their accuracy is so high for short/medium term - it is customer dependent.Lost Sales:

• There is no Lost Sales because MIBA is single source for most customers, so when MIBA cannot deliver, the order is pushed.

• Stefan's idea is that we can identify pushed orders in the  file and use it as a proxy to calculate their delivery performance. Not sure if that works, I will have a look ..

  Customer Forecast

• They have not internal method or KPI for stock-outs.