Demand-Driven MRP – Part VI: Planning
This post is part of a series about DDMRP in SAP. If you haven’t already, you are invited to read first the initial introduction and a brief explanation of why DDMRP is needed, and then the posts about the first three DDMRP components: inventory positioning, buffers, and buffer adjustments. This post is about the forth component, demand driven planning.
Demand Driven Planning
This is a big topic, so I will have to cut even more corners than usual and assume you already have some background knowledge. If you have not already, I highly encourage you to read the DDMRP book by Ptak and Smith (where the chapter devoted to Demand Driven Planning is, not surprisingly, the lengthiest one, and has several examples that will certainly make the concepts more clear).
The Demand Driven Planning component is where supply orders are actually generated. On one hand, and assuming everything else has already been aligned according to what was discussed in the previous posts, this should not be too difficult since in essence we are talking about a reorder point type planning (ok, with a few, but critical, tweaks). On the other hand, these tweaks will affect the MRP engine, the core of any ERP, so things can get tricky.
The basic idea is that you calculate a so-called “net flow position”, which is physical stock plus quantity on order minus qualified sales order demand. Sales order demand is “qualified” if it is past due, due today, or if it is a spike (and a day is considered a spike if it is a configurable number of times above ADU). If, then, the net flow position for a given part is below the top of yellow, a supply order is generated up to top of green.
Implementation in SAP
We should be able to take advantage of a lot of things from SAP’s MRP engine, but we’ll have to make some adjustments. Compared to the standard SAP reorder point planning (VB) we have to consider that:
- the reorder point itself (top of yellow) is frequently adjusted (ideally daily), mainly because of the constant adjustment of ADU;
- the replenishment quantity is also dynamic (we’ll replenish up to the top of green);
- the stock position includes elements usually not considered in a typical reorder point planning (qualified sales order demand).
Let’s deal (briefly, as always) with each of these in turn.
The reorder point must be, then, equal to top of yellow. One option would be to update the material master daily before MRP runs, but this would mean a big number of master data updates which we’ll usually want to avoid. The best option would probably be to calculate this dynamically during MRP execution (but remember ADU was already calculated beforehand) using badi MD_MRP_PARAMETERS.
We’ll want to replenish up to top of green, so lot size HB is probably the closest one to what we need. Again, the maximum stock can be updating by changing the material master or, probably better, using MD_MRP_PARAMETERS.
It should be possible to consider the correct demands in the net flow equation by configuring the mrp type (to include external requirements) and then make some adjustments / filtering of the relevant demands in badi MD_CHANGE_MRP_DATA. For dependent demands (ie, requirements from planned orders) a minor modification will be required, since the good guys at SAP decided this is not something you would want to include in reorder point planning.
We usually finish by discussing the official software compliance criteria, which in this case is:
Here a “planner workbench” is mentioned, which we have not discussed so far. This will be a list in a way similar to MD07 but adapted for DDMRP, where the planner will be able to see the situation for each material and the supply order proposals. While the development itself is not difficult, it is critical to have a good design for this transaction, as the usability of the whole solution will depend on this. Instead of using the standard MRP engine, you could of course use this workbench to create supply order suggestions. I would recommend, though, to use the MRP engine as much as possible, so that other standard functionalities are not lost (for example, lot sizing).
There is much more to Demand Driven Planning, and here, as usual, I have just covered the basics.
In the next post, I’ll cover a bit of the last DDMRP component, which deals with execution.
Hi Rui Pedro,
Thanks for the series. It is refreshing to have a clear explanation of a topic when all you read recently is marketing disguised as analysis.
There are surely advantages in having the values calculated dynamically (like buffers and their placement). The plan improvement can feel “revolutionary” when compared with the results if someone is infrequently adjusting values in the system.
However, to me the real problem with production planning (variability exists and there is no reliable way to predict it) cannot be overcome with a different way to calculate the numbers. All approaches are limited by this fact and try to improve it in similar ways.That is why when DDMRP says strategic buffers I hear safety stock and when it says Demand Adjust Factors I hear forecasting.
In any case, I find commendable that you have draft a potential SAP implementation of DDMRP with a simplified approach supported on ECC available functionality.
As you certainly know, Camelot provides a DDMRP SAP certified solution for SCM and for IBP and is “co-innovating” with SAP a Demand-Driven Rhythm Wheel Planning (DDRWP) solution. It looks to me that they all will be part of the future standard S4/HANA functionality.
But it looks like they are leaving behind customers, that would like to implement DDMRP, but are still planning with ECC. That is the niche (not that small) that you are addressing.
Let me suggest using flexible planning to enter the DAF values by periods. You can create a customized structure to both enter values as to show some calculated values as a projection of time. The values will not make sense at an aggregated level but you can save yourself all the work of coding a similar planning book transaction.
I like a good challenge and I’m dedicated to design and implement homemade solutions in APO and PP, so you can contact me directly if you want to discuss any of these ideas.
I am looking for configuration document of DDMRP with S/4 HANA.Can you help me to get the documentation or Link?
I also want this, because in many sap presentation I see that we have an app called Buffer Positioning and into this app we can go to the buffer Analysis and there, with the graph Overview, I should be able to see the net flow and the part numbers connected to each other, but so far I could not build a good example by my own.
Does anybody has a hint on it?
Very detailed description about how to achieve some of DDMRP possibly in ECC PP, still the most important part is missing there which is planning simulation. Without this function you are just doing nothing more than traditional reorder point planning and this doesn't work in combination with MRP. May I kindly ask you to elaborate on that?
Thanks & BR
Since I started writing this thread, almost two years ago, the situation has changed and now I believe the market (and SAP itself) already has SAP solutions for DDMRP. It is true those are usually S/4, IBP and / or APO (and not ECC PP), but I confess my initial motivation for writing the thread declined as those solutions started to appear.
Not sure what you mean by "planning simulation" (I can think of 2 or 3 things you might mean). The compliance criteria (www.demanddriveninstitute.com/compliant-software) does not even mention simulation, though, so it is difficult to argue that it is "the most important part".
For me two important points are indeed missing: execution follow-up (promised in the last paragraph of this post, but never actually written), and reporting / analytics.
Thanks for your reply. Planning simulation means that you will generate scheduled replenishment proposals until the end of the cumulated lead-time and not only short-term on a reactive mode. I understand this is what is behind the new MRP type D1 (Demand Driven) in S/4 HANA.
Example (without demand variability):
ADU (Average Daily Usage) = 10 PC
DLT (Decoupled Lead Time) = 100 Days
ROQ (Replenishment Order Quantity) = 200 PC
Planning Simulation will schedule 5 replenishment orders of 200 PC every 20 days
Do you understand the same or am I missing something?