Predicting Future Requirements

I have managed Materials organizations for most of my working career. For those of you who are not familiar with their organizational duties, the Materials organization within a manufacturing environment; manages the manufacturing schedule, plans and procures the components used in the manufacturing process and combines the finished products to fulfill the customer order.
The fundamental challenge that faces every Materials organization is that the total lead time to manufacture a product (e.g. component procurement and manufacturing process time) normally exceeds the amount of time your customer is willing to wait to receive the product or the lead time your competitors are willing to offer your customer. The implication is that someone, somewhere in the Materials organization has to predict future requirements, placing procurements ahead of actual customer orders.

Probably one of the best examples of this dichotomy is a fast food burger restaurant. The customer walks in, chooses a “burger combination” and expects to walk out with his/her order within a brief period of time measured in seconds. The restaurant certainly does not have time to order the raw materials or cook the meal to order.
During my career I have been asked many times why we could not put together an accurate forecast of future requirements. My response has been that “…if I could accurately predict the future, I would now be on some huge sail boat, sailing in a warm climate …”. I think that you get the gist of my thoughts.

We all know the down side of predicting requirements that the customer does not want … obsolete inventory, inventory write-offs, reduced profits!.

So what are the strategies that can make this work?? Early in the science of materials planning, the concept was to stratify the inventory into A – B – C classifications. The principle was that based upon daily dollar usage (component value times daily usage), the organization would focus on the small number of critical part numbers (e.g. A’s) and use a semi-automatic restocking strategy for the low daily dollar usage/ low cost part numbers (e.g. C’s). When the Materials group is managing 1,000’s of part numbers, there is no other choice. The disadvantages of this approach are using past usage history to predict the future, no usage history for new products/ components and a finished product can not be shipped even if it is 99.99% complete … that $0.01 part has to be there.

Current material planning science incorporates the concept of supply chain management. In essence the principle is that the greatest gain can be obtained by improving the efficiency of the entire supply chain. The entire supply chain starts from the lowest component (the farmer that raised the cow and grew the wheat in the burger example), to the next level until the final sale to the consumer (the burger restaurant). By constantly supplying usage/ consumption data to the next lower level, in near real time, that supplier has the opportunity to adjust their forecasted requirements to their manufacturing operation and suppliers. Conceptually supply chain management, reduces obsolete inventory and inventory write-offs at all levels.

So what does this have to do with globalization?? Remember my definition of globalization is “demand seeking the most efficient resource”. The word efficient does not necessarily imply lowest price! Depending on the market, efficient may mean close proximity to your customer, highly efficient Information Technology abilities, the flexibility to adjust to changing requirements without resorting to “Why can’t you provide an accurate forecast?.