DDMRP - The right math for the right problem
Over and over again I have heard comments that the math behind DDMRP is too simplistic. Yet in the real world these ‘simplistic’ formulas have resulted in my clients substantially reducing (+40%) their inventories, improving their service (+8%) and eliminating emergency costs. I have started developing a few theories on this matter.
- ''The wrong math at the wrong time'' theory
Trying to use Einstein’s theory of relativity to solve a basic quadratic equation might sound impressive but I believe even Einstein would laugh at you. All math formulas have a purpose, a time and a place, and expecting Planners to gain insights from incredible complex formulas while fighting raging operational fires is not the time nor the place. So, give the Planners the right tools that don’t require a doctorate in Applied Mathematics to interpreter and the results will surprise you.
2. "What’s the point if the ''precisely right math is just ignored'' theory?"
What has always attracted me to math, is that there is always a precise answer. However, having spent 20 years selling APS optimization solutions, it is incredibly frustrating when all this clever (and expensive) math is just ignored. One thing Demand Driven has reminded me of is that inventory operators within a range and being roughly right initially is much more powerful than being precisely wrong! ……then over time, extract the value of clever math through a continual improvement process that focuses on Order Signal Integrity and Model Outliers.
Make no mistake, I absolutely believe in clever math, At b2wise we are now using AWS’ Machine Learning to better sense demand and Right Size Inventory’s patented Monte Carlo Simulations to find a more optimal Red Zone, but only once the basic math is in place and working.
Solver of Planning Puzzles and Headaches
4 年Kevin after developing optimization models for quite some time, one premise that never fails ... "less is more". Totally agree with the concepts in your article.
Supply Chain | Systems | Process | Planning | Data | Analytics
4 年I agree with your premise entirely. In fact only the other week I came up with a very elegant solution for safety stock setting for a specific local use case and ultimately abandoned it when I realised it was just too complex. Just one thing I don’t agree with though....in the UK we spell maths with an ‘s’ :)
Inventory Management Advisor at NHS Supply Chain
4 年This is a great article Kevin and I agree that the benefit is often derived by working with the system, as opposed to trying to get the system to work with the historic operation.
SAP S/4 HANA Advanced Planning Solution Architect
4 年It is about stat not math in the case of DDMRP, think about it and figure it out
THINKFLOW and improve! Fostering Demand Driven to relieve the stress in your supply chain
4 年The main strength I found while learning the DDMRP is that it's simple logic, and simple doesn't mean simplistic, they are two different things. To optimise we need simpler solution and it looks like this is one of the way to make it simpler!