04_Dominos_Predictive_Purchase

Dominos – Predictive Purchase Order System

Project Overview

This project implements a predictive purchase order system for Domino's Pizza, forecasting future ingredient demand based on historical sales data to minimize waste and avoid stock-outs.

Objectives

• Analyze historical pizza sales data to identify demand patterns
• Build time-series forecasting models to predict future sales
• Generate automated purchase orders based on predicted demand and ingredient requirements
• Optimize inventory management

Tools & Technologies

• Python – Pandas, NumPy, Statsmodels, Prophet / ARIMA
• Scikit-learn – Machine learning models
• Matplotlib / Seaborn – Visualization
• Streamlit – Dashboard (optional)

Project Structure

04_Dominos_Predictive_Purchase/
├── README.md
├── notebooks/          # EDA, forecasting, and order generation notebooks
├── app.py              # Optional Streamlit dashboard
├── data/               # Sales and ingredient datasets
└── models/             # Saved forecasting models

Approach

1. Data Preprocessing – Clean sales data, handle missing values, engineer time features
2. Forecasting – Train ARIMA / Prophet models on pizza sales time series
3. Purchase Order Generation – Map forecasted sales to ingredient quantities
4. Evaluation – Measure forecast accuracy using MAE, RMSE

Source Repository

yasararafath-s/Dominos_predictive_purchase_order_system

📁 Project code files to be added.