Season Two, Episode Ten
The Holy Grails of Supply Chain Management
What are the things that, if we could find them, would make Supply Chain Management a whole lot easier? In fact, they are the things for which some of us search our entire careers. Have a look at my list of three. Let me know if you agree or if you have additional holy grails that you seek.
Grail #1: Knowing the “optimum” expected level (range) of inventory by part number.
There are two tasks for every inventory manager. Reduce the inventory on parts of which we have too much and increase the inventory on parts of which we have too little. We need one thing to be able to even get started. A clear and consistent method to calculate the desired inventory target for each part.
I’m sure everyone has a method for determining this and I’m equally sure that each of your methods is different. In my experience, we were always adjusting our determination of the desired inventory level. But, since many parts on the reduction list were also on my expedite list, we clearly had not found a good solution.
To determine an inventory target, we need a system that is accurate yet dynamic, connects to a range, and results in appropriate action.
Accurate yet dynamic. Every day, the inventory level and the average requirements may be rising or falling. It stands to reason that my inventory target will be higher if daily demand rises from 10 per day to 30 per day. So, every day I may need to adapt my target inventory level. Therefore, we must set up a process that automatically adapts to shifting demand.
Connects to a range. It is clearly unreasonable to expect that our target inventory and our actual on hand inventory will be identical every time. It makes sense that our expectations be tied to a range of expected inventory. If we compare our actual to a range, and track it every day, we end up with a Statistical Process Control chart for inventory per part.
Results in appropriate action. Because I can relate my on-hand balance to an inventory range, I can review the status relative to that range. If I drop below that range, I should investigate and react. If I rise above that range, I should investigate and react. If I’m getting close to the bottom of that range, I have an early warning system.
What about Safety Stock, you ask? For traditional MRP, safety stock is just the new zero. When you breach safety stock, you receive an expedite request to add inventory. This clutters your reports at best. Safety stock might save you, but at the same time, it will confuse you with an expedite request.
Demand Driven MRP elegantly handles all of these requirements. The Strategic Buffer creates a standard mechanism to estimate the expected on-hand range. And for those strategic points, a relative priority percentage can be assigned, creating that early warning system we desperately seek. By using Decoupling Points, the chaos of changes in customer demand is separated from the supplier (and vice versa). And that allows you to keep your sanity while those around you are losing theirs.
Grail #2: Knowing an accurate, consistent, and realistic lead time for production or sales.
One thing that gets overlooked or maybe taken for granted is lead time. And there are several areas where lead time is important. First, the lead time to your customer. Second, the lead time from your supplier. Third, there’s the series of internal lead times between raw stock and finished product. And fourth, any external services like heat treat, paint or powder coating that is in between your manufacturing processes.
Lead time has a significant impact on the flow of information and the expectations of everyone involved. Simply put, we depend on lead times being accurate, since we base our own plans on that accuracy. We talk about lead time as if nothing will go wrong. Because we live in a world where things do (occasionally) go wrong, the imprecision of these lead times, and the precise nature of how we use them continues to look like trying to force a square peg into a round hole.
If we had a system that managed stock positions, such that we were assured availability, then we could almost always give accurate lead times. And by monitoring the integrity of the managed stock positions (strategic decoupling points), we would have an alert that our lead time is in jeopardy.
Lead times are affected by quantity. I very seldom hear anyone talk about this. Is your lead time different if you are shipping one vs one thousand? Of course it is. How do we know when our lead time is possibly impacted by a larger order quantity? By monitoring the strategic decoupling points.
DDMRP offers this insight or visibility into the relative accuracy of our lead time promises and builds a process around communicating potential disruptions due to lead time inaccuracy or failures.
Grail #3: Knowing a problem may occur (before it occurs).
It seems self-evident that I can’t prevent a problem that has already occurred. So, I feel kind of silly even talking about it. But, most of us in Supply Chain have experienced the surprise of learning about a problem after it happens. Some of us quite frequently.
When talking about problems caused by inventory shortages, it becomes embarrassing that we aren’t aware of the impending doom. But traditional MRP is really reactionary. When an order goes into the system, we hope that the customer has given us enough lead time. Then we react. Yesterday, we had too much. Today, not enough. So, we expedite like crazy and while doing so, lose track of other issues.
A system that proactively attempts to maintain strategic inventory positions would result in less chaos, less stress and less expediting.
DDMRP fulfils this wish as well. By providing relative priority information from both a planning and an execution perspective, we have an early warning system for our ordering, our expediting, and our rescheduling.
So, if all three of these Holy Grails of Supply Chain Management are solved by an already existing methodology, then perhaps it’s time we stopped searching and started educating ourselves about Demand Driven MRP.
John Melbye, DDPP, DDLP, CSCP