Recently, I had the following discussion with a colleague regarding one of our clients. Since it was typical of many situations, I thought I would share here (without revealing any proprietary information, of course).InventoryManagementFunctions.jpg

 

------------------------------------------------------------------------

 

[Colleague] We have client that seems to have a fundamental lack of skills when it comes to inventory planning. I have had a hard time conveying the potential value some functional aspects of [their ERP system], such MRP, due to their lack of staff experience with such methods and tools. They run very successfully, but their inventory planning is a mess.


[Me] Typically, the issue is not “lack of skill,” but lack of an effective theory and appropriate strategy for inventory management. All of the tools available in [their ERP system] for inventory management can contribute to an effective inventory management system if, and only if, there is a cohesive underlying theory for modeling the inventory, and an effective strategy for managing to that model.


Most companies expect an “inventory management” system to “take over” and do the hard part for them—that is, the thinking and strategic planning for positioning inventory in their supply chain; determining how they strategically and tactically they should set system priorities; and how and when they act upon the signals the system gives them.


This is one notable advantage of [SaaS offerings today], by the way: Sage Inventory Advisor, for example, is driven by the underlying theories and strategies of its designers and developers. Unfortunately, such solutions tend to take a one-size-fits-all approach and one size simply does NOT fit all companies and industries.

 

[Colleague] Here is the basic scenario: [the Company] has only two (2) product lines—an industrial [wholesale] line and a retail line.


The [X] product is sold to industrial customers and consumed in the packing of the retail products.


For planning purposes, they are forecasting the total periodic demand for the industrial products as


Industrial Sales Forecast + Retail Consumption Requirements


Using this method, they don’t have to go thru the iterative process of MRP [material requirements planning] to drive the additional requirements for the industrial items from the forecast for the retail items.


The nature of the products means that production is done in bulk [large batches] with plant capacity constraints, so the ‘just in time’ nature of MRP is not really the best model for creating production. There is no way they could simply run MRP, create planned orders, and convert them to actual orders.


I mean: who really can? It will never reflect reality.


They are using [a process manufacturing add-on for the ERP], so any help the scheduling tool in [their ERP system] would offer is out the window.


Right now they are taking MRP up to the stage of generating the data, and then using the data in reports to assist with production planning. Sometimes, for the hell of it, they are generating planned orders as a ‘What If’ type exercise.


[Me] The biggest strength of traditional MRP is that it can explode demand through the whole bill of materials. The biggest weakness inherent in traditional MRP is that it attempts to PLAN across the bill of materials, which also means the whole cumulative lead time (from the purchase of raw materials to the completed production of finished goods). This does not really work in today’s world, so MRP sends lots of wrong signals. So many wrong signals, in fact, that managers know they cannot execute according to what MRP is telling them—at least not if they want to keep their jobs for long.


As a result, they inevitably create work-around methods to solve their “planning” problem—just like you are describing is happening in this company. These work-arounds typically take the form of Excel workbooks, paper calculations, seat-of-the-pants decisions overriding what their expensive MRP system is telling them, or custom applications developed in Microsoft Access or some other available tool.


Decoupling is the chief answer to this problem. See the attached whitepaper on decoupling.


[Colleague] The company currently has the poor business practice of creating picks ahead of time, as inventory becomes available, for firm orders to be shipped in the future. I liken this practice to seagulls fighting for a dropped French fry in a McDonald’s parking lot.


They also will produce excess inventory of the industrial items to meet future demand from retail. However, this excess product shows as available for the industrial side. I have thought through creating a bin to exclude form available, but that would exclude it from the available inventory for retail consumption requirements, too.


[Me] Strategically placed and properly sized inventory buffers should solve this entire dilemma for them. We can talk about how inventory buffer sizes should be strategically situated and sized in order to solve business problems (like decoupling lead times and absorbing variability in both supply and demand), while also maximizing return on invested capital for the inventory.


[Colleague] I think in the long run proper production planning for future demand will resolve this issue. The practice of early picking should also be retired as well.


[Me] “Proper production planning” can only be done when the signals driving the planning process are reliable and the bullwhip effect has been properly dampened by decoupling through strategically planned and implemented buffers. Otherwise, the plan that appears to be valid this morning may be wrong by three in the afternoon. And the plan that appeared to be “right on” yesterday, has to be entirely redone today. They will be fighting the same battles next week, next month or next year as they are fighting right now.


While there is much that can be done to move a company toward improvement through the application of demand-driven theory and practice, much of it would require one of three options:


      1. If they manage a relatively small number of SKU-Locations (SKULs), then, perhaps, they could get by with working out the details using Excel connected to their SQL database. I could show them how to move forward.

      2. If they have more than, say, 500 SKULs, then we could supplement [their ERP system’s] native code to create a demand-driven inventory management model within [their ERP system] that could be quite effective. The custom development would be somewhat costly, but could be done in stages.
      3. Alternatively, they could look into a couple of products that are already multi-echelon, demand-driven solutions that are ERP agnostic (using technologies similar to Sage Inventory Advisor’s export, process and present the application in a browser). These would be Replenishment+ and Orchestr8.


For your edification, I have included some other whitepapers on this topic that you might find interesting and valuable in holding this discussion with [the client].


[Colleague] There are no simple answers to this issue unless companies are willing to depart from all the methods they have already tried and found wanting (in my opinion).

 

-------------------------------------------------------------------------------------------

 

We hope you found this real-life discussion helpful. Please feel free to leave your comments below, or to contact us directly, if you prefer.

 

Follow us on Twitter: @RKLeSolutions and @RDCushing
LIKE us on Facebook: RKL eSolutions and GeeWhiz2ROI