FIGURE 4-1 The core problem of the bullwhip
If the transfer and amplification of variability in the form of demand signal distortion and supply continuity is the biggest enemy to system flow, then we have to design supply chain capability that stops or mitigates the transfer and amplification of variability through the system. But how to do that? The answer cannot be “guess better” or “eliminate all variability.” Industry has tried that for decades, spending fortunes on reengineering efforts and expensive software only to see the problem persist.
The accumulation and impact of supply and demand variability is the enemy of flow. Variability can be systematically minimized and managed, but variability will never be eliminated. The only way to stop nervousness and the bullwhip effect is to stop variation from being passed between the linked parts of the supply chain in both directions.
This is accomplished through a concept called “decoupling.” APICS defines decoupling as:
Creating independence between supply and use of material. Commonly denotes providing inventory between operations so that fluctuations in the production rate of the supplying operation do not constrain production or use rates of the next operation. (p. 43)
Decoupling disconnects one entity from another. This isolates events that happen in one entity or portion of a system and keeps them from impacting other entities or portions of the system. Think of decoupling as if it were a firewall in a building, automobile, or information system or a break wall around boats in a marina.
The concept of decoupling provides the fundamental break from convention that is needed to mitigate variability. Decoupling breaks the direct connection between dependencies. The places at which the system is decoupled are called “decoupling points.” APICS defines decoupling points as:
The locations in the product structure or distribution network where strategic inventory is placed to create independence between processes or entities. Selection of decoupling points is a strategic decision that determines customer lead times and inventory investment. (p. 43)
Decoupling is not a new idea. The concept has been around for many years but with no practical way to implement it in MRP. MRP was designed with the explicit intention of tightly coupling everything so that precise equations could be performed in order to synchronize the environment. Limited amounts of decoupling can occur in MRP, but only with complications where costs likely outweigh their benefits (this is discussed further in Chapter 9).
Figure 4-2 is based on Figure 3-3 and depicts the dependent view of an MRP system and the accumulated demand signal distortion (the upper arrow moving right to left) and the supply continuity variability (the lower arrow moving left to right). There is no decoupling; thus the distortion to both relevant information and materials accumulates through the system.
FIGURE 4-2 The MRP approach
Decoupling points represent a place to disconnect the events happening on one side from the events happening on the other side. They delineate the boundaries of independently planned and managed horizons. The determination of their placement is not to be taken lightly—it is a strategic decision that will dramatically affect how the system operates and how effective the overall system will be.
For the decoupling points to maintain their decoupling effect, there must be a level of protection that absorbs demand and supply variability at the same time. This level of protection is a concept called “decoupling inventory.” APICS defines decoupling inventory as:
An amount of inventory kept between entities in a manufacturing or distribution network to create independence between processes or entities. The objective of decoupling inventory is to disconnect the rate of use from the rate of supply of the item. (p. 43)
Decoupling point inventory in this book will be referred to as a “decoupling point buffer” or simply “buffer.” Decoupling point buffers are quantities of inventory or stock that are designed to decouple demand from supply. Buffers are commonly amounts of inventory that will provide reliable availability to the consumers of the stock while at the same time allowing for the aggregation of demand orders, creating a more stable, realistic and efficient supply signal to suppliers of that stock.
Figure 4-3 depicts the same system as Figure 4-2 but with decoupling point buffers in place. The placement of decoupling point buffers (represented as the tiered bucket icons in the dependent structure) creates independent planning and execution horizons. These horizons are indicated by the dotted lines with rounded terminal points on each side. Demand and supply variability is stopped from further accumulation at those terminal points. This is represented by the wall-like icons labeled “break wall.” This means that the use of decoupling point buffers addresses both components of the bullwhip at the same time and from the same place; it is a bidirectional solution.
Decoupling buffer placement has huge implications for lead times. By decoupling supplying lead times from the consumption side of the buffer, lead times are instantly compressed between buffers and to the customer. This lead time compression has immediate service and inventory implications. Market opportunities can be exploited, while working capital required in buffers placed at higher levels in the product structure can be minimized.
Furthermore, Figure 4-3 reveals an additional lead time compression benefit due to decoupling: its impact on relevant information. As discussed in Chapter 3, MRP users are forced to make commitments to a demand signal that is subject to varying degrees of error (forecasted orders). While there is an alternative much more accurate demand signal (sales orders), MRP’s inability to decouple prevents the exclusive use of that signal.
FIGURE 4-3 A system with decoupling point buffers
Yet what Figure 4-3 shows is that when a decoupling point buffer is placed inside the sales order visibility horizon, it will allow for the system to exclusively use that accurate demand input. We have effectively found the time that we believed we lacked that forced the use of forecasted orders in the first place. When decoupling point buffers match the sales order visibility horizon, the demand variability is reduced.
Decoupling simply makes sense given the basic circumstances that we face today. We have extended supply chains globally and made them more complex and fragile. These longer and more complex supply chains are subject to much higher levels of variability and are much harder to plan. Breaking dependencies in key places will dramatically simplify the planning equation and provides