Quantum of inventory buffer change in supply chains is key to maintaining a balance between stability and responsiveness.
We have discussed in the last three posts how inventory buffers should be sized, what should be the trigger for buffer changes and how frequently we should resize the buffers. Now we come to the most critical aspect of the dynamic buffer management. What should be the quantum of buffer change?
If the quantum is high, the supply side may not have the capability to respond to a big change, leading to instability in flow. On the other hand, if the quantum is too small, we may not be responsive to the demand changes. Where is the sweet spot?
Since buffers are initially set at the maximum demand during RLT, this definition should hold good in future as well. Now that we have access to demand patterns at the granular level, we can predict the short-term demand at that level using AI/ML, and computing resources are no longer a constraint, we can very well recalculate the buffer sizes every time.
Conventional practice is to revise the buffers upward or downward by 33%, which may be the reason for causing instability in less flexible supply chains. Does 33% have a scientific basis? I think it is time to challenge this practice and make revisions which match the extent of shift in demand patterns.
Keeping the quantum of buffer change flexible helps in achieving both the objectives of providing stability and being responsive to market demand. It also takes away the imaginary fear that frequent buffer revisions would lead to instability.