In an outstanding whitepaper released by Tata Consultancy Services, Lakshmi Narasimhan and Ramesh Srinivasan discuss a topic—among other things—that certainly needs to be clarified.
The term “demand driven” has been applied to supply chains with such a broad range of realities underlying them that it has been difficult for the real demand-driven supply chain to separate itself from the confusion.
In their whitepaper, Narasimhan and Srinivasan cover five common misconceptions. Drawing upon their writing, and adding what I hope will be helpful comments, I will try to cover these misconceptions here:
1. Demand-driven means make-to-order
While it seems like a fairly rational leap from “truly demand-driven” to “make-to-order,” the leap is too great. When we use the term “truly demand-driven,” we mean to express a supply chain where replenishment actions are always triggered by actual demand, and not from forecasts. In a truly demand-driven environment, strategically placed buffers function as decoupling point in the supply chain. The purpose of these buffers is to absorb variability in supply and demand. However, it is actual demand that brings any given buffer to the point where replenishment actions are triggered. (Note: Forecasts related to future events may influence the size and configuration of any given buffer, when necessary, but forecasts never directly result in replenishment actions in a truly demand-driven supply chain.)
The replenishment actions are not triggered by specific customer orders—as in make-to-order. Rather, they are triggered by and prioritized based on the status of the strategic buffer.
2. Decoupling points in a demand-driven supply chain mark the push-pull boundary
In a truly demand-driven supply chain, “push” generally does not come into play. The decoupling points in a truly demand-driven supply chain serve two purposes: 1) to decouple lead times and bring them within customer tolerances, and 2) to reduce or eliminate the transmission of variability across the supply-demand boundary (i.e., bullwhip effect). “Push” does not happen on either side of the boundary under demand-driven methodologies. (Note: As in many other things, purity is seldom the full reality. Even in environments that operate as truly demand-driven over 99+ percent of the SKU-locations involved, there may be a few that simply cannot be handled in this way. As a result, there may be a small number of items still handled in a “push” scenario.)
3. Demand-driven systems never employ forecasts
In fact, truly demand-driven environments may rely on forecasts in more than one way. Two crucial role of forecasts and forecasting in truly demand-driven supply chains: 1) long-term forecasting – which is used primarily for capacity planning into the future, driving capital investment and other longer-term decisions; and 2) short-term forecasting – used to calculate buffer adjustment factors which, when applied, scale buffers up and down to cover “events” such as short-term promotions, product introductions and end-of-life, or other events such as adding new stores or distribution channels.
4. Demand-driven is all about placing burdens on suppliers to stay “stocked up”
On the contrary, truly demand-driven environments are genuinely focused on being able to supply better, more accurate data about demand to suppliers while, at the same time, reducing variability in demand. The goal of truly demand-driven supply chain manager should be to increase end-to-end visibility and to aid suppliers in their own movement toward becoming demand-driven. It’s good for everyone in the supply chain.
5. Demand-driven is only good for raw materials and low-value goods
Narasimhan and Srinivasan say, “Many companies implement pull replenishment mechanisms only for C-class (less critical or low value) parts or parts that experience steady demand.” However, as a result of this approach, these companies miss the opportunities that demand-driven methodologies offer for improving the return-on-investment performance of their critical A and B-class items, as well. With few exceptions that are typically easily identified and handled, becoming truly demand-driven is good for the whole gamut of inventories across the whole span of the supply chain.
The damaging effects of these misconceptions
Companies and supply chain managers who felt that they could not become MTO (make-to-order) or could not give up forecasting may have turned aside from becoming truly demand-driven due to these or other common misconceptions. We think it is high time that demand-driven MRP and demand-driven S&OP be given a second look—even, a deep look.