Skip to content

Amneetkaur24/pricing-inventory-optimization

Repository files navigation

Pricing & Inventory Optimization

Demand Elasticity Modeling • Revenue Simulation • Inventory-Aware Pricing

image

Project Overview

Retail and e-commerce companies often optimize prices using theoretical demand models, without considering inventory constraints.

This leads to:

• Stockouts during promotions • Lost revenue opportunities • Inefficient discount strategies • Misalignment between pricing and operations

This project builds an end-to-end pricing analytics framework that evaluates whether pricing strategies are operationally feasible, not just theoretically profitable.

The analysis integrates price elasticity modeling, scenario simulation, and inventory constraints to identify pricing strategies that maximize realized revenue.


Business Question

Can a pricing strategy that maximizes demand actually be fulfilled with available inventory?

The project evaluates how pricing decisions interact with:

• demand elasticity • revenue potential • inventory availability • stockout risk


Interactive dashboard highlights:

• Pricing scenario KPIs • Theoretical vs feasible revenue • Category-level pricing recommendations • Elasticity vs inventory risk analysis

Live App https://pricing-inventory-optimization-2appeg9cg3dxgr9u6utkhjx.streamlit.app/


Project Workflow

1️⃣ Data Preparation

• Cleaned raw pricing and demand data • Handled missing values and inconsistencies • Created analysis-ready datasets


2️⃣ Price Elasticity Analysis

Analyzed how demand responds to price changes.

Identified:

• Elastic product categories • Inelastic product categories • promotion-sensitive segments


3️⃣ Demand Modeling

A regression model was trained to estimate demand at different price points.

This allowed simulation of revenue curves across pricing ranges.


4️⃣ Pricing Scenario Simulation

Simulated multiple pricing strategies:

• Baseline price • −10% discount • +5% price increase • +10% price increase

Each scenario evaluates:

• predicted demand • projected revenue • demand sensitivity


5️⃣ Inventory-Aware Pricing Evaluation

Inventory constraints were applied to determine feasible sales volume.

Key metrics evaluated:

• fulfilled demand • stockout risk • realized revenue

This step highlights the difference between theoretical revenue and achievable revenue.


6️⃣ Elasticity & Risk Segmentation

Categories were segmented based on:

• price elasticity • stockout probability

This enables identification of:

• safe discount ranges • high-risk promotional strategies


Key Insights

• Deep discounts increase demand but significantly raise stockout risk • Inventory constraints reduce achievable revenue versus theoretical forecasts • Moderate pricing strategies deliver the most reliable revenue outcomes • Pricing decisions must align with supply planning


Business Recommendations

Pricing Strategy

Avoid aggressive discounts without inventory readiness.

Moderate pricing adjustments produce more stable revenue outcomes.


Inventory Planning

Promotional campaigns should be aligned with:

• inventory availability • replenishment cycles • demand forecasts


Executive Takeaway

The most profitable pricing strategy is not always the one that maximizes demand — it is the one that maximizes fulfilled revenue.


Tools & Technologies

Programming

• Python • Pandas • NumPy

Modeling

• Linear Regression (Demand Modeling) • Price Elasticity Analysis

Visualization

• Matplotlib

Dashboard

• Streamlit

Version Control

• Git & GitHub


Project Structure

pricing-inventory-optimization
│
├── 01_app
│   ├── app_pricing_and_inventory_optimization.ipynb
│   └── app_pricing_and_inventory_optimization.py
│
├── 02_data
│   ├── category_decision_table.csv
│   ├── ecommerce_pricing_cleaned.csv
│   ├── ecommerce_pricing_featured.csv
│   ├── ecommerce_pricing_featured_02.csv
│   ├── ecommerce_pricing_raw.csv
│   ├── elasticity_risk_table.csv
│   └── scenario_summary.csv
│
├── 03_notebooks
│   ├── 01_Category_Strategy.ipynb.ipynb
│   ├── 02_Demand_Modeling.ipynb.ipynb
│   ├── 03_Elasticity_Analysis.ipynb.ipynb
│   ├── 04_Elasticity_Risk_Segmentation.ipynb.ipynb
│   ├── 05_Inventory_Feasibility.ipynb.ipynb
│   ├── 06_Pricing_Scenarios.ipynb.ipynb
│   └── 07_Final Summary Notebook.ipynb.ipynb
│
├── 04_insights.md
├── 05_Dashboard Price And Inventory.pbix
├── 06_Pricing and Optimization.pptx
├── README.md
└── requirements.txt

Assumptions

• Inventory levels are simulated • Demand model uses linear regression for interpretability • Results illustrate pricing strategy trade-offs

Production deployment would require:

• real inventory feeds • advanced elasticity models • continuous monitoring


Future Enhancements

• Non-linear demand modeling • Dynamic pricing optimization • Supply-chain lead time integration • Explainable AI for price decisions • Automated price recommendation engine


Author

Amneet Kaur Data Analyst | Pricing Analytics | Inventory Optimization Canada