Tampering: Demand & Inventory Planning

Last week, we posted a blog re-introducing Deming’s concept of tampering.  This week we will explore the how tampering creeps into demand and inventory planning.

In the case of demand planning, tampering occurs when we simply think they are smarter than a well crafted statistical forecast.  A statistical forecast uses past history and gives more emphasis to recent performance to better capture trends.  People from any and every function think they are smarter than all this history and the mathematical manipulation thereof.  We can always rationalize and even fabricate reasons why we will sell more or less… usually more.  There are stories in almost every business sector about statistical forecasts being overlaid by the so called hockey stick where future sales will suddenly spike in the next few months or quarters.

Management will often disparage the planning team for poor forecasts.  They will challenge the team to improve forecasting accuracy but not giving them any help in reducing the variation, the common cause variation, responsible for forecast accuracy being so low.  To improve on a system with only common cause variation requires structural change.  In the realm of demand planning, the structural change is usually associated with rigorous SKU management.  

When it comes to inventory management there are several common cause factors that may lead to what management deems as excessive inventory.  There are three such factors.  Lead time has become the largest factor in inventory sizing.  With so many products or components of product being imported from Asia, the physics of inventory sizing demand at least inventory plus some safety stock.  If the lead time is eight weeks, there must be at least eight weeks of inventory in the system.  Forcing something less is ill advised from a service standpoint and is tampering.  The second factor is sales fluctuations.  Sales fluctuations or variation influences the size of the safety stock.  If the business experiences high degrees of month and quarter end peaking, these peaks have to be reflected in the inventory sizing. Forcing inventory to be less than that dictated by lead times and sales variations is tampering and will result in out of stocks.

A well executed and managed Sales and Operations Planning (S&OP) can help in a few key ways.

1. Good data management:  Lead times and sales histories are clean and up to date.  This will improve inventory balance and allow the team to optimize to the common cause sources of variation.


2. If there is a need to further reduce inventory, the cross functional nature of S&OP will have Sales, Marketing, Finance, and the Supply Chain all being aware of what could be done to make reductions:
   
   
   1. Reducing Lead times
   
   
   2. Dampening Sales Variations:  this includes both
      -  Month End Peaking
      -  Variation due to SKU proliferation
   
   

These last two are harder things to solve.  Variation due to SKU proliferation can be mitigated with a sound and effective portfolio management process which is usually implemented in tandem or as a result of S&OP.  Dampening month and quarter end peaking is a much harder issue to address.  Customers have been trained or trained us to believe that they will delay purchases until there is such pressure for us to make sales numbers that we offer discounts and incentives that customers cannot ignore.

What are your experiences in tampering in Demand and Inventory Planning and Management?
Have you had success reducing month and quarter end peaking?  If so, what did you do?