Dynamic Pricing Decision Simulator

A decision-support simulator for evaluating pricing strategies under demand uncertainty, capacity constraints, and customer risk.

Why This Project Exists

Dynamic pricing in volatile markets is often treated as a pure optimization problem. However, real-world implementations face significant challenges that pure optimization misses:

• Static pricing under volatile demand leads to missed revenue and inventory imbalances.
• Need for pre-deployment strategy evaluation: Leaders need to know "what if" before committing to a pricing policy.
• Trade-offs: Maximizing revenue often comes at the cost of customer satisfaction or operational stress.

This project creates a safe, simulation-based environment to stress-test pricing strategies against realistic market conditions before they touch production systems.

What This Project Does

• Forecasts short-horizon demand using historical booking data.
• Applies governance-first pricing policies, prioritizing business rules and constraints over black-box optimization.
• Simulates outcomes under capacity constraints, modeling sell-outs and spoilage.
• Compares strategies using business metrics, providing clear scorecards on Revenue, Load Factor, and Booking Curve health.

What This Project Is NOT

• Not real-time pricing: This is a strategic planning tool, not a live pricing engine.
• Not autonomous optimization: The system recommends and evaluates; it does not auto-update production prices.
• Not production deployment: This repository contains the simulation and decision support logic, decoupled from transactional systems.

Architecture Overview

graph LR
    Data[Historical Data] --> Forecast[Demand Forecast]
    Forecast --> Policy[Pricing Policy]
    Policy --> Simulation[Market Simulation]
    Simulation --> Scorecards[Performance Metrics]
    Scorecards --> Dashboard[Executive Dashboard]

Loading

How to Run

To run the full end-to-end pipeline (simulation -> evaluation -> dashboard):

npm install
npm run full

This command will:

1. Install dependencies.
2. Run the Python simulation pipeline (generating data, running policies, evaluating results).
3. Launch the Next.js executive dashboard.

Demo Walkthrough

For a comprehensive review of the system capabilities, we recommend visiting the following views in order:

1. Dashboard Home: High-level comparison of the Baseline vs. Recommended strategies.
2. Simulation Details: Deep dive into specific booking curves and daily performance.
3. Policy Configuration: Review the specific logic and parameters driving the recommended strategy.

Key Result

In backtesting against high-volatility periods, the POLICY_RECOMMENDED strategy demonstrates:

• ~4–5% Revenue Lift compared to static pricing.
• Lower Operational Stress by smoothing demand peaks.
• Improved Load Factors without aggressive last-minute discounting.

Tech Stack

• Python: Core simulation logic (pandas, numpy).
• Next.js (App Router): Executive dashboard and interactive visualizations.
• Tailwind CSS: Rapid, responsive UI development.
• Recharts: Data visualization for booking curves and revenue metrics.