Introduction
Once a volume axis enters your inventory, the next thing you'll want to do is rank the SKUs, don't you think? Staring at every SKU uniformly gives you no clue where to start.
So we combine two axes — turnover and storage cost — to classify SKUs into ranks A/B/C/D. The aim is to cleanly separate "stars that sell well and take little space" from "stock that doesn't sell and hogs space." This article walks through the concrete calculation design of that ranking.
The Idea Behind Rank Classification
Evaluating Inventory on Two Axes
Traditional ABC analysis usually lines up SKUs along a single axis such as sales or inventory value. For warehouse optimization, that wasn't enough. What I use are two axes: turnover (how much it moves) and storage cost (how much space it eats). Fast movement plus good space efficiency is ideal; slow movement that only eats space is a problem child. Capturing inventory in these two dimensions reveals its character in three-dimensional relief.
Separating the "Stars" from the "Problem Children"
The essence of rank classification is deciding where to direct limited attention. SKUs that sell well and take little space (the stars) contribute to the warehouse even if left alone. Conversely, SKUs that don't sell and occupy large volume (the problem children) are the top candidates for intervention. Ranking is a yardstick that mechanically sorts this "inventory to praise" from "inventory to act on."
How to Calculate Storage-Efficiency Cost
Theoretical Minimum Cost from Demand × Volume
The core of storage cost is the product of demand (shipment history) and volume. We estimate how much space × time an SKU theoretically needs to turn over, and treat that as its "theoretical minimum storage cost." For a fast-moving item, space rotates quickly so the theoretical cost is low; for a dead item, the same space is occupied for a long time so the theoretical cost is high. This becomes the baseline for efficiency evaluation.
Measuring the "Real Weight" with an Allocation Multiplier
Against that theoretical minimum, the allocation multiplier expresses how much extra space the actual inventory eats. By seeing how many times the theoretical footprint an SKU occupies, you can pinpoint SKUs that "should turn over in about this much space, yet take up far too much room." The higher the multiplier, the heavier the burden that problem child places on the warehouse.
Prepare shipment history and computed volume per SKU
Compute theoretical storage cost from demand × volume
Derive how many times the theoretical footprint is actually occupied
Combine turnover and storage cost, classify at A/B/C/D boundaries
Setting the Thresholds
Where you draw the A/B/C/D boundaries shapes how usable the analysis is. Cutting every SKU by absolute values distorts the distribution due to seasonality and product-mix bias. So I base it on relative thresholds — top N percent as A, bottom N percent as D — while dropping obvious problem children to D via absolute conditions too. Leaving room to fine-tune the boundaries during operation proved practical.
Ranks are not fixed
Demand fluctuates by season, and volume-estimation accuracy improves over time. Ranks aren't decided once and done — design them on the premise of periodic recalculation and updates.
Aggregating by Location
Viewing by Fixture, Block, and Location
Once you have per-SKU ranks, aggregate them by fixture, block, and location. Then mismatches between placement and inventory character emerge: "this fixture is full of rank-D problem children," "this prime spot is occupied by rank C." Beyond individual SKUs, taking a bird's-eye view of space as clusters makes it easier to spot where a relayout will pay off.
Connecting to the Next Action
Rank classification and location aggregation aren't the goal in themselves. Stars to easy-to-pick spots, problem children to disposal or order adjustment — value appears only when this connects to concrete action. How to translate it is covered in detail in the related article "Turning Analysis into Shelf-Layout and Disposal Decisions."
Summary
By classifying into A/B/C/D on the two axes of turnover and storage cost, you can mechanically separate "stars that sell well and take little space" from "problem children that don't sell and hog space." The core of the calculation was the theoretical minimum cost from demand × volume, and the allocation multiplier that measures actual occupancy.
Once the ranks are out, it's time to turn them into action. Head to "Turning Analysis into Shelf-Layout and Disposal Decisions." For volume-calculation details see "Why Add 'Volume' as an Inventory Analysis Axis," and for the big picture, the hub article.