May 17, 2017
Lesson 4: Setting Inventory Policy
Setting inventory policy in line with realistic targets
“If you don’t know where you are going, any road will take you there.”
Setting objectives and goals for your inventory is fundamental to the success of your business – we call that Inventory Policy.
Inventory Policy is all about balancing two things:
- Investment in inventory
- Service level to your customers (sometimes called Fill Rate)
Conventional wisdom would have you believe that the better service level you want to provide to your customers, the more money you have to invest in inventory in your warehouse.
On the other hand, if you want to save some money in your warehouse, your service level to customers will almost certainly suffer.
But what if you could get better overall service levels to your customers while also saving money invested in inventory? Recall the first lesson in this course when you classified your inventory. Well, if you invest your money into the A items and take investment away from the C items, you can achieve your goals. This is because you don’t have to stock as much of the inventory your customers don’t buy from you. But you have to invest more into the inventory that your customers actually buy. Sounds simple, but how do you define a stocking policy?
Defining a stocking policy
The only practical way to set a stocking policy is to define these two things:
- What service level you want to provide your customers
- What kind of replenishment or order cycle you want to use
You may remember from a previous day in this course that these are both inputs to the Safety Stock calculation. That’s right: the best way to control your stocking policy is through setting the correct Safety Stock.
And to get to the goal of achieving better service levels to your customers whilst saving money, you have to set a different stocking policy for items that are classified differently. Here’s an example stocking policy:
- A items: service level target of 97%, replenishment cycle of 2 weeks
- B items: service level target of 92%, replenishment cycle of 4 weeks
- C items: service level target of 85%, replenishment cycle of 12 weeks
What does that mean?
Let’s take the A items: we’re striving for a pretty high service level of 97%. That means that when a customer orders an A item, we want to be able to supply that customer immediately 97% of the time. Be careful with setting this too high, there’s an exponential function of service level vs money invested here. Just one percentage point higher may cost you as much as double the inventory to achieve! Note that you can never aim for a 100% service level because that would mean an infinitely larger warehouse.
The replenishment cycle of 2 weeks means that whenever we order A items from our supplier, we want to order 2 weeks worth of forecasted sales. This means that we will place frequent smaller orders. This is a good idea because it keeps the cash flow requirements low. Even if your supply Lead Time is more than two weeks, you can still have more than one order “in flight” at any given time. The point where making the replenishment cycle too short comes when other costs (like admin, shipping, etc.) start dominating the overall order cost.
Now look at the C items: these are the complete opposite. A low service level here is OK because these items are in low demand and if you lose the sale of one of these, you won’t lose a high-profit sale. Why the large replenishment cycle? Well, if you’re going to bother reviewing the order requirements for C items, you might as well buy a bunch of them and forget about it for a while. The only time you’d want to shrink the replenishment cycle for those items is if they are large and hard to warehouse or because of cash-flow constraints.
Using these two inputs combined with every item’s risk factors, you can calculate a Safety Stock level for each item. The Safety Stock will be your ideal minimum level of stock. The Safety Stock + Replenishment Cycle level will be your maximum level. And halfway between min and max lies your model stock level.
If you sum the model stock levels for every item in the warehouse, you’ll arrive at the total ideal investment in inventory for the whole warehouse. If that number is too expensive for you, you can start making the Target Service Levels lower to reach a target inventory value. At least now you know exactly what the consequences of your decisions are.
Let’s get practical
Does all the above make sense but sound impossible to do? With the right tools at your disposal, this it is a lot easier. However, if the only tool you have available now is a spreadsheet, you’ll need some tips and ideas to make this all work.
Measure actual fill rate
Of course, tracking this daily with every order is the best answer, but let’s start with something to give us an idea of where we are now per group:
- Load your items with their current net stock (physical stock less any customer back orders that should have been shipped already) per warehouse into your spreadsheet.
- Carry forward the groups and the ABC classification you did earlier into separate sheets here.
- Create a simple “VLookup” formula to get this all onto one data sheet so you can see what the net stock is per product with the products ABC classification, average forecast and unit cost.
- Create a potential forecast value per item as the average forecast multiplied by the unit cost.
- Create a little “if statement” that tests if the stock is zero or less and makes that a zero, and if the stock is greater than 0, make that a 1 – this is really a Boolean indicator to show if the item has stock or not. You can get fancier with this later, but this is a good starting point.
- Take this stock-out flag and multiply it by the value of the average forecast to get a weighted forecast result. Add that up per ABC (or other groupings if you want) and divide by the potential forecast value to get an achieved fill rate per group.
It’s advisable to do this analysis each week for a while to determine if this is getting better or worse using the policy parameters you are using for ordering. If the result is not as great as you wanted, then start considering higher factors in that safety stock calculation we touched on earlier in the course.
Value of the orders
It may not just be the result of previous orders that you want to consider here. You see when we amend these parameters, we change the next order. In one of the following lessons, we will talk about calculating those ideal levels from these inputs and calculating an order per product. It is important to consider whether we can actually afford to buy against the policy set. So make sure to check the resulting orders created by the policy you set here and see that this can be afforded now. For now, consider smaller replenishment cycles for the products and maybe lower service levels so that you can gradually improve the result in your business at a rate you can afford.