# Product Segmentation for Retail with Python

A statistical methodology to segment your products based on turnover and demand variability

A statistical methodology to segment your products based on turnover and demand variability

** Product segmentation **refers to the activity of

**and**

**grouping products that have similar characteristics****. It is usually related to marketing**

**serve a similar market***or manufacturing*

*(Sales Categories)**.*

*(Production Processes)*However, ** as a Logistics Manager**, you rarely care about the product itself when managing goods flows;

*except for the dangerous and oversized products.*Your attention is mainly focused on the ** sales volumes distribution** (fast/slow movers),

**and**

**demand variability****.**

**delivery lead time**You want to put efforts into managing products that have:

- The highest contribution to your total turnover:
**ABC Analysis** - The most unstable demand:
**Demand Variability**

In this article, we will introduce simple ** statistical tools to combine ABC Analysis and Demand Variability **to perform

**products segmentation.**

**SUMMARY**

**I. Scenario**1. Problem Statement

2. Scope Analysis

3. Objective

**II. Segmentation**ABC Analysis

Demand Stability: Coefficient of Variation

Normality Test

**III. Conclusion**## I. Scenario

1. Problem Statement

You are the Operational Director of a local ** Distribution Center (DC)** that delivers

**.**

**10 Hypermarkets**In your scope you the responsibility of

- Preparation and delivery of replenishment orders from stores
- Demand Planning and Inventory Management

*You can find the full code in this Github repository:**Link*### 2. Scope Analysis

This analysis will be based on the M5 Forecasting dataset of Walmart stores' sales records (Link).

We suppose that we only have the first-year data (d_1 to d_365):

in**10 stores****3 states (USA)**unique SKU**1,878**and**3 categories****7 departments (sub-category)**

Except for the warehouse layout, categories and departments have no impact on your ordering, picking or shipping processes.

**Code â€” Data Processing**

### 3. Objective

What does impact your logistic performance?

**Products Rotation***What are the references that are driving most of your sales?*

- Very Fast Movers: top 5% (Class A)
- The following 15% of fast movers (Class B)
- The remaining 80% of very slow movers (Class C)

This classification will impact,

**Warehouse Layout**

Reduce Warehouse Space with the Pareto Principle using Python

**Picking Process**

Improve Warehouse Productivity using Order Batching with Python

**Demand Variability***How stable is your customersâ€™ demand?*

- Average Sales: Âµ
- Standard Deviation:
- Coefficient of Variation: CV = Ïƒ/Âµ

For SKUs with ** a high value of CV**, you may face

**that would lead to workload peaks, forecasting complexity and stock-outs.**

**unstable customer demand**Code

- Filter on the first year of sales for HOBBIES Skus
- Calculate the Mean, Standard deviation and CV of sales
- Sorting (Descending) and Cumulative sales calculation for ABC analysis

## II. Product Segmentation

This analysis will be done for the SKU in the HOBBIES category.

### 1. ABC Analysis

What are the references that are driving most of your sales?

**Class A: the top 5%**- Number of SKU: 16

- Turnover (%): 25%

**Class B: the following 15%**- Number of SKU: 48

- Turnover (%): 31%

**Class C: the 80% slow movers**- Number of SKU: 253

- Turnover (%): 43%

In this example, we cannot clearly observe the Pareto Law ** (20% of SKU making 80% of the turnover)**.

However, we still have 80% of our portfolio making less than 50% of the sales.

**Code**

### 2. Demand Stability: Coefficient of Variation

*How stable is your customersâ€™ demand?*

From the Logistics Manager's point of view, it is way more challenging to handle a peak of sales than a uniform distribution throughout the year.

In order to understand which products will bring planning and distribution challenges, we will compute the coefficient of variation of the yearly distribution of sales of each reference.

**Class A**Fortunately, most of the A SKUs have a quite stable demand;

We won't be challenged by the most important SKUs.

**Class B**The majority of SKUs are in the stable area;

However, we still spend effort on ensuring optimal planning for the few references that have a high CV.

**Class C**Most of the SKUs have a high value of CV;

For this kind of reference, a cause analysis would provide better results than a statistical approach for forecasting.

**Code**

## 3. Normality Test

*Can we assume that the sales follow a normal distribution?*

Most of the simple inventory management methods are based on the assumption that the demand follows a normal distribution.

**Why?**

Because itâ€™s easy.

**Sanity Check**

Before starting to implement rules and perform forecasts itâ€™s better to verify that this hypothesis cannot be refuted.

Weâ€™ll be using the** Shapiro-Wilk test** for normality; it can be implemented using the Scipy library. The null hypothesis will be

*.*

*(H0: the demand sales follow a normal distribution)***Bad News**

For an alpha = 0.05, we can reject the null hypothesis for most of the SKUs. This will impact the complexity of inventory management assumptions.

**Code**

## III. Conclusion

This operationally driven segmentation gives us a few insights into the challenges your operations will face for planning and managing the goods flows to meet your store's demand.

## References

[1] Reduce Warehouse Space with the Pareto Principle using Python

[2] Improve Warehouse Productivity using Order Batching with Python

[3] Scipy stats Shapiro Test documentation, Link