A Demand Driven Buffer, what’s that?
In the Demand Driven jargon we talk about “buffers” of different types. What are they used for? How do they work? This post introduces some basic concepts, which are explored in detail in the Demand Driven Institute’s training courses.
Buffers and variability.
If you have any experience in supply chain you know that daily life is full of surprises: the machine is down, a component is short, the supplier is late because it’s overloaded, an urgent order has just come in, a batch is blocked for quality, a key operator is ill – and sometimes all of these things all at once. Does that sound familiar?
Whatever Lean 6 Sigma efforts are made to reduce variability, it is and will remain pervasive. To ensure the continuity of our flows and the agility of our responses to the market, we need to position dampers in our system that will allow us to absorb some of the variability.
Trying to precisely synchronize all our operations, no matter how sophisticated the technology used, is a red herring. We are living in an imperfect world full of surprises
To dampen the variability, we have 3 possibilities:
- We can use stock. This is often the easiest to implement. This is the heart of the DDMRP, typically the first step in setting up a Demand Driven approach. It applies very well from the moment there is a recurring demand for candidate items for storage.
- We can insert time. In other words, organize queues in our streams. This tactic is particularly well suited to make to order processes: if your items are non-recurring, you can’t store them and you need time to buffer variability.
- We may keep available capacity. If certain process steps are “overcapacities” you can absorb peaks in demand or catch up with a delay due to a contingency.
These three types of buffers complement each other and “collaborate” to build a robust model: if you have excess capacity you need less stock and can reduce your downstream queues.
You’ll tell me, this isn’t all new, and it’s all common sense. Right, and yet…
Inserting stock, time (and therefore delay), available capacities: isn’t all this contrary to Lean, which advocates the elimination of waste?
Beware of misinterpretations of Lean! Lean consists in focusing on the value brought to the customers, i.e. on the flow. The founding fathers of Toyota’s production system, from which Lean originated, never said that stock was a waste – they said overstock is a waste!… The right stock, the right time, the right available capacity are investments to ensure focus on the flow and therefore on the value delivered to customers!
Take control: Positioning and sizing to protect flow.
Stock, queues, available capacities waiting to be loaded: all factories and distribution centers experience that! But are these stocks, queues, or available capacities undergone or under control?
The first step in setting up a Demand Driven model is to design the model : the flow from one end of the industrial system to the other is described, analyzed, and a decision is made on where to position the stock, time and capacity buffers, how to size them, and how to drive them. These are investment decisions to build a resilient and agile supply chain.
It is a question of investing wisely: the buffers of stocks, time and capacity are positioned at only a few strategic points, which will make it possible to steer priorities.
The process is quite like future state value stream mapping (VSM).
It uses data analysis of the existing model, but also and above all the collective intelligence of the teams.
Use the buffers to take action! 
A buffer will trigger decisions by our teams on a daily basis. To do this, whether it is a question of stock, time or capacity, we set up a management system based on red / yellow / green color codes. This visualization allows a shared understanding of priorities, and decision making that promotes a fast and reliable flow.
Before being named DDMRP the model was called “actively synchronized replenishment”. It is all about giving teams the signals to act: place an order, speed up an order in progress, add staff to a station, slow down a means to avoid clogging a queue, etc.
Using buffers to improve.
Each buffer is a data collection point. Daily the events on the buffers are archived: was the stock red/yellow/green/blue? Was the flow equation red/yellow/green/blue? Was the reception in the time buffer in the red/yellow/green/blue zone? How many days were our capacity buffers loaded in red/yellow/green/blue?
In view of these analytics, how can we improve the performance of our model, i.e. improve its speed, reliability and stability while reducing investments?
How do I do that in my ERP?
Most ERPs on the market do not integrate the logic necessary to drive a complete Demand Driven operating model, including the management of stock, time, and capacity buffers. Most of the solutions available on the market only deal with stock buffers. Demand Driven Technologies’ solutions allow the implementation of all three types of buffers within any ERP environment... Do not hesitate to contact us for more information!
