Are you losing sleep at night worrying about forecast accuracy? Here’s something that can help.
The absolute measure of forecast accuracy is, simply, whether or not your product is on the shelf – where and when the customer wants it – and not before. Everyone – including your sales people who live by the availability rule that “You can’t sell from an empty cart” – knows that it’s impossible to forecast demand that precisely at that level.
Isaac Asimov’s famous Foundation series illustrates this in a true, science-based science fiction story. In it, Asimov presents the concept of “psychohistory,” a set of mathematical formulae that can predict the future of a society. However, the developer of “psychohistory” emphasizes that it only works for large social segments – it cannot predict the behavior of any individual.
Similarly, formal, mathematical forecasting of product demand is more accurate at higher levels of aggregation – even with collaboration. It cannot predict the behavior of any individual customer.
So, what do we do for insurance against the inevitable inaccuracy of any forecast?
One of the things we can do is employ global available-to-promise (GATP).
Spreading product across many shelves increases the likelihood of it being on some shelf, somewhere in our supply chain, when our customer wants it. And GATP, simply put, provides visibility into where it is through a feature called location substitution.
Providing products that are acceptable substitutes is another way that we can increase the likelihood of satisfying our customer’s need. GATP provides another feature called product substitution: if our customer’s first choice isn’t available anywhere in our supply chain, GATP gives us other, similar products that are acceptable to him. Location substitution and product substitution can work in conjunction with each other and can substantially improve our customer service level and overcome the uncertainty and imprecision of our forecasts.
Finally, at any point in time, the next best thing to giving our customer what he wants when he wants it, is to give him accurate information about when he can get it. GATP has precise information not only about a product’s stock level and location, but also about its total supply and demand picture. The APO system – through its real-time integration with the transaction system – knows exactly what purchase requisitions, purchase orders, stock transfer requisitions and orders, planned and actual in-process production orders, and any other factors constituting supply. So, by including this information in its scope of check functionality, GATP can give our service representatives and our customer precise information about when his requirement can be filled.
So, GATP relies on actual, up-to-the-minute – up-to-the-second, really – transactions that are fed by input from all the supply side departments in the organization. While this data may not be perfectly accurate itself, it is still closer to reality than intuition, insight, and mathematical formulae. It is the best reflection of collaboration.
So, what I’ve talked about here are the key elements of GATP:
- Location substitution
- Product substitution
- Scope of check, and
- Real-time integration.
Naturally, GATP is going to work most effectively for companies that can make use of these organizational and business process circumstances, such as distributing their inventory across multiple warehouses. But there are a couple of other key considerations that any company considering implementing GATP must take into account. The first is that the company’s policy elevates customer service above the cost of doing business. Simply, delivering product from a more remote location will affect transportation cost. Secondly, products must be stocked commonly in at least several, if not most of its distributed warehouses.
Thirdly, as we’ve come to know from decades of computer-based MRP systems, not only inventory accuracy, but indeed the accuracy of all transactions recorded in “The System,” is critical. A customer’s tolerance of variances in on-time delivery will be tested if you’re off because of sloppy house-keeping about your purchase orders’ expected receipt dates.
Obviously, we need forecasting more than ever in today’s global business environment. But my whole point is: let’s make the best of both worlds – aggregate, long range planning, and insuring the best possible on-the-shelf availability.
In Asimov’s day it was science fiction; in today’s business, it’s survival.