I had my first involvement with DDMRP back in 2015. The requirement came up during the design phase of a large transformation program for a global business. I have a long industry background in Manufacturing rooted in SAP MRP and during those 15 years, I never once had been in a situation where I thought there was any functionality missing, so I was initially very sceptical that there was anything new to be found when I was asked about DDMRP.
The project I was running was yet another SAP implementation and so we didn’t have the option to use DDMRP in the standard planning tools available but after several discussions with the client, we began to see that there could in fact be some benefits for products where the forecast was so inaccurate that you could actually believe they had used the legendary “blindfold and dartboard method”. SAP were not interested in DDMRP at the time, so we enabled DDMRP planning in SAP IBP ourselves using custom key figures. After the long slog to achieve this, I now see SAP have agreed this method is worthy of inclusion in the standard product (well done and thanks to all involved) and it is now included in core S/4HANA and standard SAP Integrated Business Planning functions. Perhaps we were too far ahead of the curve, but having built this functionality from scratch, I have seen how things work from the ground up and believe it can be useful for planning materials that met a particular criterion (more on that later).
During the build in 2015, the original question I had stilled remained. Why was this any different to Safety Stock Planning? If you listen to any DDMRP evangelist, DDMRP is the silver bullet cure for the ancient MRP style planning. Quite hard to believe when MRP has stood the test of time and even in today’s world of customisable products with “segments and batches of one”, I still think there is room and a need for both of these approaches. The key decision point is product segmentation.
Decoupling is one of the things you hear so much about with DDMRP, but with the correct approach to master data, this was already achievable in MRP. So, for an experienced MRP person, there is “NOTHING TO SEE HEAR” when you hear talk about decoupling. “Bright Canning” and other multi-echelon stocking policies have been around well before the DDMRP drum was being banged. When used properly, safety stocks are deployed at all BOM Levels to cope with forecast variations at finished product level. No Bullwhip here if I’m holding the right safety stocks at the semi-finished goods stage and even better for materials that were generic across multiple finished products. Value Adding Services in the Warehouse or during freight, with labelling or product customisation at the last possible moment, has been a Good Manufacturing Practice for many years.
The key thing that DDMRP brings to decoupling is the mindset needed in considering how and where to decouple; DDMRP has a defined process in place to cover this, but sorry to say this is old news.
Where DDMRP comes firmly into play is where products are difficult to forecast or have high variability in demand or supply. HOWEVER, here comes the big point to consider. DDMRP is agnostic of the forecast errors because it only considers actual demand. Hence the term Demand Driven MRP. This is where the segmentation and the specific criterion I mentioned earlier comes into play. I have worked in a number of FMCG businesses where the lead time on orders is significantly less than the lead time for the making or buying the product. In this situation I don’t see how the forecast can be ignored and as a result DDMRP is not applicable for every product you have, hence the need for segmentation to start the process. This is because the yellow buffer zone alone will cover one lead time of the product using an average daily usage. In some business I have worked in, that amount of stock alone would quadruple stock holding before adding in the Green and Red Buffers.
So, the clear message is; find the right replenishment strategy for every component, semi-finished and Finished Product you have and use tools to do the segmentation automatically on a scheduled basis in line with your S&OP cycle. Identify where you can decouple, and then replenish using the method that best suits the product. Dynamic re-order point planning can still work very well in improving service levels on minimal stocks when set up correctly.
I like what DDMRP has done to the market. It has given us all the shake-up we needed in today’s 2-Hour Delivery World with omni-channel requirements. There is definitely a place for DDMRP for some products in your portfolio and there are benefits from all five steps (or three steps depending on your view). I prefer the three-step thinking because most people can relate to groups of three things. Just think Position – Protect – Pull.
If you think in this way, there is still room for old fashioned Re-order point planning for bulk items or MRP PD strategies for stable core products with short order cycles, as long as the material master data has been set up correctly. In my experience, adjusting MRP settings according to segmentation rules can often bring about huge inventory savings without the need for additional investment.
So, does DDMRP mark the end for MRP? Definitely not both can work together to provide appropriate plans for varying requirements across your product range. It is more like the Tonic in a G&T, bringing a bit of spark and complementing the MRP processes that were already in place very well. For longer shelf life items with comparable replenishment lead times, it is definitely worth taking a look and seeing how DDMRP could eliminate uncertainty from Multi-Echelon Inventory Planning. The goal for all these methods being improved service levels without hiking up the amount of inventory you need to hold. But I would say don’t throw away MRP just yet! More important is segmenting your products correctly to provide the best fit replenishment strategy across your portfolio.
Comments