Inventory management involves balancing the tradeoff of stock-outs (a customer service measure) versus inventory control (a cost-control measure). Optimization is the mathematical and computer methodology that allows the evaluation of trillions of choices for inventory levels to occur very quickly and find 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
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 see if each SKU is trending up or down, or has seasonal sales patterns. These algorithms are optimized to maximize forecast accuracy.
Forecasting is one aspect of optimizing inventory levels, because it looks forward to anticipated sales. The second aspect is safety stock, which protects against stock outs. Safety stock looks backwards at how well the forecasting algorithms were able to anticipate demand. Highly accurate forecasts need very little inventory. However, SKUs with high forecast error need more safety stock. Supply Velocity’s Inventory Optimizer uses forecast error, the item’s lead time and profitability to let each SKU set its optimal inventory level to maximize the overall profit-return-on-inventory-investment.
Other Inventory Control considerations include how many customers purchase and how many times customers order each SKU. These factors will be used to modify the mathematical optimization’s output to balance achieving the highest service levels for the most customers while controlling your inventory investment.
One important input to the optimization algorithm is inventory carrying 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 criterion is obsolescence. For food companies such as cheese producers, this cost is very high because of short shelf-lives. For hard-goods companies, such as machine-tools, this cost is very low because their parts never go out of date or rust. 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 inventory carrying costs between 10% and 45%.
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 manufacturing cellular production, supply chain network design, supply chain performance measurement and inventory optimization.
Inventory Optimization – Outline
- Perform inventory-network analysis
- If you have more than one warehouse, this step determines what items should be stored in which locations.
- Gather inventory data
- SKU demand, lead-time, gross profit, number of customers
- Calculate inventory carrying costs
- Run forecast and safety stock optimization model
- Make adjustments based on additional criteria
- Re-run monthly