Inventory management involves balancing the tradeoff of stock-outs, a Voice of the Customer performance measure, versus inventory control, a Voice of the Business cost-control performance measure. Inventory optimization is one of the most controllable and important aspects of supply chain management.
Inventory Optimization uses a computer math modeling that allows for the very swift evaluation of trillions of choices for inventory levels, in order to determine the most profitable inventory level for every item (or stock keeping unit, SKU) that you stock and sell. Inventory Optimization includes proven inventory management techniques and mathematical optimization algorithms to maximize your profit-return-on-inventory-investment. These methods build on the popular ABC inventory control method, but use the latest mathematical and computer technology to achieve optimal results.
Steps to Optimize Inventory
Demand Forecasting
Any company that carries inventory is forecasting, whether they think so or not. Many companies claim that they don’t forecast, but if you are stocking raw materials, work-in-process or finished goods in anticipation of fulfilling demand, you are forecasting. At Supply Velocity, we run multiple, proven forecast algorithms to determine if each SKU is trending up or down, and/or if there is a seasonality factor. These algorithms are optimized to maximize forecast accuracy.
ABC Inventory Classification, Safety Stock and Service Levels
ABC Analysis or SKU Classification is an important first step in inventory planning. To differentiate SKUs, and the customer service level (or expected fill rate) applied to safety stocks, we utilize multi-criteria inventory classification (MCIC). Multi-criteria inventory classification allows the organization to determine what criteria defines a high performing SKU for their business. Often, we see organizations choose either profit, sales, or volume, but these criteria may not provide a holistic picture. Using multiple criteria specific to your organization, the MCIC model determines high and low performing SKUs for your business. The lowest performing SKUs will not be stocked and will only be produced or replenished on-demand. The remaining SKUs will be stocked based on their ABC Classification with greater service levels and safety stock for A SKUs and less for C or D SKUs.
Demand Forecasting is an important component of inventory optimization because it “looks” forward to anticipated customer demand. Another aspect is safety stock, which protects against stock outs due to variation. Safety stock calculations look backwards at how well the forecasting algorithms were able to anticipate demand. Highly accurate forecasts need very little safety stock while SKUs with high forecast error need more safety stock. Supply Velocity’s inventory optimization solution uses forecast error, the item lead time, SKU classification and service level goals to set each SKU’s optimal inventory level to maximize the overall profit-return-on-inventory-investment of the company.
Inventory Holding Cost
An important input to the inventory optimization algorithm is inventory holding cost. This is different for every company and is based on three criteria. The first is the cost of borrowing, or cost of money. Since the year 2000, this has been very low, with most companies being able to borrow at 5% or less. The second is obsolescence, often caused by holding excess inventory. For food companies such as cheese producers, and the entire perishable food supply chain, this cost is very high because of short shelf-lives. For hard-goods companies such as machine-tools, and OEM supply chains, this cost is very low because their parts only slowly go “out of date” or deteriorate. The final cost is storage. If your products are very small and don’t require temperature control, this cost is very low. We have calculated that inventory carrying costs in supply chains can vary between 10% and 45% of revenue.
Target Service Levels
The next step in inventory optimization is determining your target service levels. This will allow you to choose the tradeoff between inventory costs (impacting working capital) and your inventory targets or service levels. The higher you set your service levels the more inventory you will hold, and more demand you can fulfill from stock. However, your inventory costs will increase. Another consideration is the cost of a stock-out. In hospital supply chains the cost of a stock-out may be critical, including the loss of life. Service levels and safety stock levels in these supply chains are very high. In retail supply chains the cost is possibly the loss of that sale or maybe the loss of the customer forever. Different supply chain strategies and stock-out costs will inform how you set your safety stock service levels.
Reorder Points, Reorder Quantities and your ERP System
The final step in inventory optimization or inventory management is connecting the demand forecast with your SKU classification, lead time and target service levels, to determine your SKU level inventory targets. These targets could be managed in your ERP system through MIN / MAX levels or Reorder Point / Reorder Quantity.
Monthly Updating of Inventory Levels
To support your supply chain and ensure robust inventory management, the forecasting process and calculations of reorder points and order quantities, should be updated monthly. This will ensure your inventory strategy is based on the most up to date customer demand and that your inventory levels support your supply chain strategy. Inventory optimization also supports a supply chain planning process. To keep your supply chain operating efficiently, it is important to have the right inventory and also to achieve high inventory turns.
In addition, with monthly updates your replenishment team can trust their inventory levels and you can react to changes in your supply chain such as increases or decreases in lead times.
Our founder, Mitch Millstein, Ph.D. is a widely recognized researcher and teacher in the area of Supply Chain Optimization, Supply Chain Analysis and Supply Chain Decisions. His research includes ecommerce and omnichannel supply chains, supply chain network design, supply chain performance measurement, inventory management, and inventory optimization.