Someone told me…

The original vision was that the DDMRP – methodology for managing flows and stocks – did not need forecasts to work. The reality is a bit more nuanced. Answers with Jérémy Catteloin, expert engineer in supply chain management at BEVOLTA.

“In a DDMRP tool, the calculation of the need is based on the orders in portfolio. However, the stock level objectives – that is to say the level of the buffers – can possibly be calculated on the basis of a smoothed forecast over a certain horizon (or not, depending on the configuration mode of the buffer in question) .

We can therefore rely on or completely dispense with forecasts, this point is generally arbitrated in the project phase. We can come back to it later if necessary. The most important thing is that we use the forecasts where they are least false, in particular to size the stock targets, layer by layer (in the buffer), over a given time horizon, without going to the nearest comma, being finally quite robust with respect to the small forecast errors that can be had from one week to the next or from one day to the next. And stay fair overall.

In the DDMRP, the calculation of today's need is based on the orders in the portfolio today, while in the classic MRP, the calculation of today's need is based on the forecast that was made there. has some time on this need. We immediately understand the largest margin of error of the MRP system!

In the DDMRP we will calculate the CMJ – the average daily consumption – in a smoother way. We will take into account the reality of the orders of the week to calculate the needs for this week, where in the MRP we induce a stronger variability, which becomes more and more false as we go up in the nomenclature. What works on a finished product turns out to be completely wrong upstream of the nomenclature.

By smoothing, we bring stability with the DDMRP methodology. This is key for an industrial system. This avoids the Christmas wreath effect with permanent “stop & go” and a stronger “bullwhip effect”.

With the buffers we attenuate, we absorb it, in short, we better control the variability of the Supply Chain. »

Interview by Ghislain Journé