Inventory Analysis Part 2: Item Profiles and Order Profiles, the keys to everything.

ABC analysis.

I use ABC categories in a lot of the item profile analyses. I’ve written a separate article on ABC/Pareto, and I suggest you read that first.

Summary versus Detail Profiles.

Item profiles and order profiles can be presented in both summary and detail forms. For the purposes of this article, I’m focusing on the summary profiles. The summary profiles are used for higher-level decision-making, such as determining order-picking methods or slotting logic. The detail profiles are used to execute a plan. For example, the summary information may show you that you have 500 very fast movers that will take up one pick aisle (a fast-pick aisle); the detail would apply those same characteristics to actual items so you can do the detailed slotting.
I think the best way to start is to show an example of an item profiles table.

Item Profiles

Below is an example of a very basic summarized Item Profile table. It’s broken down by ABC (ABC by times sold) and shows the number of SKUs and Number of Picks in each group.

ItemProfile

The ABC Description provides some context for each ABC classification with A items being picked/sold greater than 5 times per day. That makes it easier for others to understand how you classified ABC. I also calculate average picks per day (by dividing picks per year by 250) to also make it easier to understand. The Cumulative Percentage of SKUs and Cumulative Percentage of Picks are important because they show the level of the Pareto Principle for these SKUs. You can see the combined A and B items account for 81% of the picks, but only 13% of the SKUs. You can also see that the slowest movers (E items) include 43% of the SKUs but only result in 1% of the picks. This is typical of what I see. Unfortunately, everyone has a lot of slow movers.

Order Profile Table.

Below is an example of a very basic order profile table.

OrderProfile

This table focuses on the line items (picks) per order. This type of breakdown would be used for things like analysis of order-picking methodology. Overall, they have an average of around 3 lines per order. But you can see that doesn’t accurately reflect what they are actually processing.
The Line-Item groups I break down the data into, are based on what I see in the data and how I may use the profiles. In this example, they have very few orders greater than 15 lines, so I made several groups below that. Some businesses will have a much wider range of order sizes. If they had orders that were greater than 100 lines, I would break out that group because it’s useful to know that. In this group, the largest order was only about 25 lines, so I stopped at the >15 group. 1-line orders are unique in that there are picking methods that work well just for these types of orders, so as long as there are a significant number of single-line orders, I will always list those separately. Note that single-line orders account for 41% of orders but only 13% of line items. This is important because even if you get a picking process to efficiently pick those orders, you still have a lot of picks on other orders to process.

Other Stuff

These are very basic examples. It’s important to consider how you intend to be using this information. You may need to expand on these tables or create other tables based on other characteristics. For example, I mentioned that single-line orders are rather unique, but single-line-single-piece are a unique subset of that group, and if you have a lot of these, you may want to break that down further. Also, you may want to incorporate cube information in these tables or create separate tables based on cube. Order profiles based on Cube-per-order in evaluating conveyor systems, picking carts, material handling equipment, or order-packing methods. Cubic Velocity (the total cube that goes through a specific slot over a week, month, year) can be useful in slotting decisions. Non-Conveyables are another category you may need to break out. This would be important if you use conveyor or any other equipment that may have issues with large or heavy items.

The examples here are for small items (piece-pick), but if you handle some full pallets, you would probably want to include some information on that. I use the term Full-Pallet-Equivalent (FPE) frequently when I don’t have data available to say how much is on a pallet by item. I will just convert cube to FPE in the tables based on whatever this company would consider a full pallet.

You’ll notice I didn’t incorporate pieces-per-pick in the examples. I find that is not a significant characteristic for many operations. However, if it does have a significant impact, I will either incorporate that in the existing tables or create a separate one to represent that characteristic. Similarly, if scale counts are an issue in an operation, I will see if there is a way to identify picks that are likely scale counts.

There are a lot of averages used in the examples, but sometimes that isn’t adequate for evaluation. If an operation is highly seasonal, or has significant variability during the week or even at specific times of the day, I will try to quantify that. For example, a lot of e-commerce operations are very busy on Mondays. You need to plan for those peak periods.

Storage Profiles.

This is another category of a profile you may need to consider putting together. Just about every warehouse is or will be short of space at some point. A storage profile not only helps to quantify this, but also helps when evaluating storage options. This would be especially important with full-pallet storage. You may categorize by pallet size, then list the number of SKUs and pallets currently in stock. Optionally you can just use cube and calculate Full-Pallet-Equivelents. It gets trickier in piece-pick operations.

Annual Analysis

Several decades ago, I started doing annual analysis of operational data for the organization I was responsible for at the time. This can be very useful for making apples-to-apples comparisons if things change in the organization. If your lines per order or pieces per pick or picks per day changes, that is going to have an effect on your processes. It doesn’t help to just say, “I think we’re busier”; you need to know what has changed so you can make the best decisions.

I’ll again mention that the way you profile your items and orders depends on the intended use of those profiles and those characteristics that have an impact on decisions you may be making.