Inventory Optimisation: Finding the Right Balance

Inventory is one of the most misunderstood assets in supply chain management. It appears on the balance sheet as a current asset, which implies value. But holding inventory is expensive -- capital tied up in stock earns no return, warehouse space and handling add cost, and the risk of obsolescence, damage, or loss grows with every day an item sits on the shelf. The cost of holding inventory is typically estimated at between 20 and 30 per cent of its value per year when all the carrying costs are included.

At the same time, not holding enough inventory carries its own costs: lost sales, emergency procurement at premium prices, production stoppages, customer service failures, and the reputational damage that comes from repeatedly being unable to meet demand. The cost of a stockout can easily exceed the cost of holding the inventory that would have prevented it.

The discipline of inventory optimisation is not about minimising inventory as an end in itself -- that approach trades one set of costs for another. It is about finding the minimum level of inventory that achieves a defined service standard, given the actual demand and supply variability the organisation faces. That is a more precise and more useful goal -- but it requires understanding the different types of inventory and the levers that determine each.

The Three Types of Inventory

Not all inventory serves the same purpose, and the approach to optimising each type is different. Understanding which type of inventory you are looking at is the first step in deciding how to manage it.

• Cycle stock is the inventory that is consumed and replenished in the normal course of business. It is driven by order quantities: if you order in batches of 500 units and consume 100 units per week, your average cycle stock is 250 units. Cycle stock can be reduced by reducing order quantities -- which reduces the average level held at any point in time. The constraint is that smaller order quantities typically mean higher ordering costs and, in some cases, loss of volume discounts. Economic Order Quantity (EOQ) analysis finds the order quantity that minimises the combined cost of holding and ordering.

• Safety stock is the buffer held to protect against variability -- in demand, in supplier lead times, or in both. It is the inventory that sits above the cycle stock level and is not expected to be consumed in normal operations. Safety stock exists because forecasts are never perfect and lead times are never completely predictable. The appropriate level of safety stock depends on the variability of demand, the variability of supply lead time, and the service level the organisation is committed to achieving.

• Anticipation inventory is built ahead of a known future demand event -- a seasonal peak, a planned promotion, a period when the supplier will be unavailable (for maintenance, shutdown, or capacity reallocation). Unlike safety stock, anticipation inventory is planned and time-limited: it is built before a known event and drawn down during or after it. The key questions are whether the event is genuinely predictable and whether the cost of building the anticipation inventory is less than the alternative (expediting, lost sales, or temporary sourcing from a higher-cost supplier).

Calculating the Right Safety Stock Level

Safety stock is the type of inventory that is most frequently either over-held or under-held, and the most amenable to disciplined calculation. The standard formula for safety stock is: safety stock = Z × σ_LT × D, where Z is the service level factor (drawn from a normal distribution table for the target service level), σ_LT is the standard deviation of demand during lead time, and D is the average daily demand.

In practice, the most important inputs to get right are the service level target (which determines Z) and the lead time demand variability (which is a function of both demand variability and lead time variability). A common mistake is to use average lead time and average demand in the calculation without accounting for the variability of either. An item with highly variable lead times and moderately variable demand requires significantly more safety stock than the average-only calculation suggests -- and the stockouts that result from under-stocking it will confirm this expensively.

It is also worth noting that service level targets should be set deliberately rather than defaulted to. A 99 per cent service level sounds reassuring but requires dramatically more safety stock than a 95 per cent service level -- and the question of whether that additional investment is justified depends on what a stockout actually costs in each specific context. High-margin, high-demand items with large stockout costs justify high service level targets. Low-margin, easily substitutable items with low stockout costs do not.

The ABC-XYZ Matrix for Prioritising Inventory Management Effort

ABC analysis ranks inventory items by their contribution to total inventory value (A items are the top 20 per cent of items that typically represent 80 per cent of value; B items are the middle tier; C items are the long tail of low-value items). XYZ analysis adds a second dimension: demand predictability. X items have stable, predictable demand; Y items have moderate demand variability; Z items have highly erratic or intermittent demand.

Combining the two analyses produces a nine-cell matrix that guides where to direct inventory management effort and which management approach fits each segment. AX items (high value, predictable demand) are the prime targets for tight cycle stock management and precise safety stock calculation. AZ items (high value, erratic demand) are the ones that keep inventory planners awake at night: the high value argues for low stock levels, but the erratic demand makes any level feel insufficient. BZ and CZ items are often candidates for Kanban-style replenishment rather than formal safety stock calculation, because the demand pattern makes statistical methods unreliable.

Common Optimisation Mistakes

The most damaging inventory optimisation mistake is cutting safety stock across the board as a cash-release exercise. When an organisation reduces inventory by applying a uniform percentage reduction to all safety stock levels, it does not reduce the highest-risk holdings first -- it reduces everything proportionally, including the safety stock that was already at a minimum and the safety stock on items where a stockout is genuinely expensive. Service levels deteriorate, stockouts increase, and the cash savings are offset by emergency procurement costs and lost margin.

The second common mistake is ignoring lead time variability in safety stock calculations. Many organisations calculate safety stock using average lead time and average demand, which produces a number that is correct on average but insufficient when lead times extend beyond their average. If a supplier quotes a four-week lead time but occasionally delivers in six or seven weeks, the safety stock calculation that assumes four weeks will regularly result in stockouts during those extended cycles.

The third mistake is treating inventory optimisation as a one-time project. Demand patterns shift, suppliers change, service level commitments are renegotiated, and the variability parameters that determined last year's safety stock levels may not reflect this year's reality. Inventory policy should be reviewed at least annually -- and immediately when a significant change in demand pattern, supplier performance, or business strategy occurs.

XNM Consulting helps organisations optimise inventory policies, improve demand planning accuracy, and build supply chain practices that balance service performance with working capital efficiency. Learn more about our procurement, sourcing, and contract management services.