Beauty Products Supply Chain Data Analysis and Visualization

This project involves an in-depth analysis of a supply chain dataset focused on beauty products to uncover insights, identify trends, and visualize various aspects of the data. The primary objectives are to understand customer preferences, supplier performance, and key factors impacting different metrics within the supply chain.

Table of Contents

• Overview
• Data Preparation
• Correlation Analysis
• Data Visualization
• Feature Importance Analysis
• Key Insights
• Usage

Overview

The project focuses on the analysis of a beauty products supply chain dataset to extract meaningful insights that can help in decision-making and optimization within supply chain management. Various visualizations are used to represent the findings, and a Decision Tree Regressor model is employed to identify the most impactful features on product cost and manufacturing lead time.

Data Preparation

1. Loading the Data: The dataset is loaded using pandas.
2. Initial Exploration:
   • Display the first few rows and descriptive statistics.
   • Check for missing values and confirm their absence.
   • Encode categorical variables using one-hot encoding for further analysis.

Correlation Analysis

1. Pearson Correlation:
   • Calculated to check for linear relationships between features.
   • Filtered values to identify significant correlations.
2. Spearman Correlation:
   • Calculated to check for monotonic relationships between features.
   • Filtered values to identify significant monotonic relationships.

Data Visualization

1. Product Quantities by Location:

   • Bar chart showing product quantities sold by different locations.

2. Distribution of Product Types by Customer Demographics:

   • Pie charts displaying the distribution of product types among various customer demographics.

3. Defect Rates by Supplier:

   • Bar chart comparing average defect rates among different suppliers.

4. Order Quantities by Routes and Transportation Modes:

   • Stacked bar chart visualizing the distribution of order quantities across different routes segmented by transportation modes.

5. Average Lead Time by Supplier and Product Type:

   • Bar chart analyzing the average lead time for each supplier by product type.

6. Manufacturing Costs by Product Type:

   • Box plots showing the distribution of manufacturing costs for different product types.

7. Product Costs by Product Type:

   • Box plots showing the distribution of product costs for different product types.

8. Shipping Costs vs. Order Quantities:

   • Scatter plots exploring the relationship between shipping costs and order quantities for different carriers.

9. Revenue Generated by Customer Demographics:

   • Pie chart showing the revenue generated by each customer demographic.

Feature Importance Analysis

Using a Decision Tree Regressor to identify the top features that impact:

1. Product Costs:

   • Determined the top 5 features affecting the cost of products.

2. Manufacturing Lead Time:

   • Determined the top 5 features affecting the manufacturing lead time.

Key Insights

• Customer Preferences: Female customers primarily prefer skincare and cosmetic products, while male customers favor haircare products.
• Supplier Performance: Supplier 5 has the highest average defect rates, whereas Supplier 1 has the lowest.
• Order Distribution: Different transportation modes affect order quantities across various routes.
• Lead Time Analysis: Average lead time varies significantly by supplier and product type.
• Cost Analysis: Key features impacting product costs and manufacturing lead time identified using the Decision Tree Regressor.

Usage

Clone the repository:

git clone https://github.com/mishra-krishna/Analysis-and-Optimization-of-Supply-Chain-Operations
cd Analysis-and-Optimization-of-Supply-Chain-Operations
pip install -r requirements.txt
Beauty_Products_Supply_Chain_Analysis.ipynb