Optimizing Supply Chain Flow: Beyond Inventory Management

Numerous software solutions, and not the least, are available to optimize inventories. Just Google “inventory optimization” and you’ll see a wide range of tools aimed at tackling stock management.

In fact, if we take a closer look, a stock can’t be optimized, let’s see why.

First of all, you need to understand what stock is and what supply chain modeling entails.

Let’s look at stock from two angles:

• Stock is an investment in supply chain modeling. Stock will be positioned at a given location to ensure availability at that stage. For example, to ensure 24-hour dispatch on an e-commerce site, you’ll need a stock of finished products.
• Stock is a consequence, a symptom resulting from planning decisions, actual demand, actual receipts, a combination of decisions, and the consequences of events that result in your having 5 pallets of this reference in your distribution center today.

Let’s now consider the intrinsically dynamic nature of a supply chain. Markets generate demand, i.e., the rate at which your products are consumed, which leads upstream to the rate at which your semi-finished products and purchased components are consumed, and which propagates through your network of suppliers in the form of the rate at which multiple items are consumed.

This rate of consumption – the rhythm of demand – is supported for each item by a supply chain flow.

It’s the gap between the speed at which this supply arrives and the speed at which the market consumes that causes the stock to rise or fall. At the extreme, consumption disappears and obsolete stock is generated – or consumption is higher and we generate shortages. Which stops consumption.

So what we need to optimize are flows, not inventories. Flow optimization means timing the flow of supplies as closely as possible to actual demand. When demand increases, we pull in more supplies, and when demand decreases, we slow down.

It’s about adjusting the throughput of a flow, as opposed to filling static reservoirs. This approach is essential for dynamic supply chain planning.

To enable this continuous adaptation of supply to demand, we need to break a number of bad habits.

Demand-driven tactics take us by the hand to help us:

• A DDMRP buffer does not describe a stock, but a flow. The stacking of red, yellow, and green zones describes the total “net flow” to be maintained.
• We are constantly identifying the pace of demand. This item has an average historical consumption of 12 per day and an average forecast of 14 per day. This is very different from a “next month’s forecast is 280 and the following month’s is 195” approach. We’re not thinking in terms of discrete quantities per period but in terms of a continuous rhythm of demand and adaptation of the supply rhythm.
• We combat distortions in the rhythm of demand. If the green zone is too high, it generates infrequent replenishment, which disconnects the rhythm of consumption and supply – this is obvious and we try to reduce its impact.
• We reduce the fixed horizons as much as possible, which, by definition, disconnects the speed of consumption and supply. Smoothing, yes – freezing, no – or as late as possible.
• At regular intervals, we adjust the design of the flow model – decoupling points, delays, time buffers, capacity constraints – so that the operating model enables a dynamic supply chain planning and reliable response to demand.
• We give ourselves S&OP visibility on an ongoing basis and make the necessary arbitrages at the right moment rather than next month.

For more information on optimizing supply chain flows, visit our Intuiflow page.

Flow optimization is very different from the historical tactics of forecast optimization and safety stock optimization. Intuiflow uses analytical techniques (Auto-Pilot AI, process mining, simulation, etc.), but above all, it draws on the collective knowledge and intelligence of the teams to design the supply chain flow model and continuously improve it.