The supply chain revolves around supply and demand. If suppliers delivered on time every time and we could predict what our customers were going to buy, we could match the supply and demand, and all would be well in the world. Unfortunately, life doesn’t work like that.
Suppliers lead times often vary, and they may not always provide you with the full shipment in one delivery. Additionally, customer demand fluctuates as a result of seasonal or fashionable trends. In the real world, suppliers are not always reliable, and inventory forecasts are not always right. To protect yourself from demand and supply risks, you need a certain amount of safety stock in place. This will ensure that your customer service levels remain consistently high.
In our personal lives, we have car insurance to mitigate the risk should something go wrong. We are happy to pay for a policy that we may never use so that we can sleep better at night knowing we are financially protected. There are risk profiles associated with car insurance policies, and your monthly premium depends on factors such as your age, demographics, and previous history. The same applies to factors when looking at safety stock. Insurance companies look at the history and behavior of the driver. Inventory planners look at the history and behavior of a stock item to determine what its safety ‘premium’ should be.
Safety stock is there to protect you from late or part deliveries from suppliers and from underestimating your sales, so you can sleep better at night knowing your inventory is covered.
There are a few factors and considerations to take into account when calculating what your safety stock should be.
Analyze the data on your supplier’s order delivery history.
- Look at your order information against your goods received information and specifically look for:
- how often the items were delivered in full or only part.
- supplier anomalies – strip these out of your equation so as not to skew the results.
- Irregularities such as goods that were airfreighted in on special orders or orders where there was an unexpected delay.
- By taking out these anomalies, you get a more accurate representation of how that supplier generally performs. Look at how well the supplier delivers stock to you in terms of both quantity and time.
Sales & previous forecast history
From an outbound supply chain perspective, analyze your sales history against your previous forecast. Look at your level of accuracy and predictability – what is the variation between sales and forecast?
- It is critical that you don’t just look at the last forecast attempt for the month compared to the sales – the purchase decision is often taken months in advance of that.
- Make sure to start the measure when there was stock of the item to sell rather than the date the product was added to the master list.
- Factor in both computer and staff overrides into the forecast measure – it is the combination that results in what you end up buying.
By factoring in the real risk associated with the forecast accuracies into the calculation of your safety stock, you will have the buffer you need to help improve buying and sales teams to do what they do best.
Target fill rate
We would all aim to provide 100% customer service levels by being in stock 100% of the time. Unfortunately, this is not always possible. The target fill rate on your fast movers needs to be higher than your slow movers – you don’t want to be caught without stock on our popular items! The higher the target fill rate of the item, the more safety stock is held to buffer the risk. Remember to model that impact. Setting high target service levels when you have poor service right now, will cost a lot of extra stock – can you afford to do that all in one go? Should you rather slowly increase the levels to get to where you want to be?
Calculate your safety stock in days, not units.
Some systems and tools out there allow you to set the safety stock or minimum stock in units. If you are using units, for example, 100 units in high season may not be enough, but in the low season may be way too much. Instead, use a time metric for this. Using days, for example, 25 days’ worth of inventory may be 200 units, but in low season 25 days may only be 10 units. In this way, you can bring your inventory down but, at the same time, improve your service level to your customers.
To keep your fill rates up, without burying your working capital in excess stock, keep analyzing the risks, and adjust your safety stock accordingly. Long lead times and bad forecasts are the root of all evil when it comes to inventory. Work on the business processes and tools you use when looking at suppliers and forecasting to really get some amazing results.