ContinuumSC: Continuous Operations Platform

ContinuumSC is an end-to-end MLOps intelligence system designed to optimize supply chain operations through continuous machine learning. The platform automates demand forecasting and inventory optimization using a sleek, interactive command center.

Features

• Demand forecasting — XGBoost with time-series feature engineering, predicting future product demand from historical trends, seasonality and lags.
• Inventory optimization — calculates Safety Stock and Reorder Points using standard service-level formulas.
• Interactive dashboard — a modern, dark-themed Streamlit UI with clear visualisations and operational controls.
• Custom data ingestion — upload your own CSV or Excel supply chain data, or use the built-in generator to try the platform with simulated data.

Installation

1. Clone the repository

   git clone https://github.com/SaadTadja/continuumsc.git
   cd continuumsc

2. Install dependencies

   pip install -r requirements.txt

3. Run the application

   streamlit run app.py

No configuration is required for a first run — if you do not upload a file, the built-in generator produces a simulated dataset so the dashboard is populated immediately.

Custom data schema

If you upload your own data via the Upload CSV/Excel sidebar tool, it should contain the following columns:

Column	Type	Meaning
Date	datetime / string	Timestamp of the record
Product_ID	string	Unique product identifier
Demand	integer / float	Daily demand volume
Lead_Time	integer / float	Lead time to replenish stock
Unit_Cost	float	Cost per unit
Holding_Cost_Rate	float	Share of unit cost to hold inventory

Project structure

app.py                    main Streamlit dashboard
src/engine.py             forecasting and optimization logic (XGBoost)
src/data_generator.py     simulated data when no file is provided
src/styles.py             custom UI styling
.streamlit/config.toml    theme configuration

License

MIT — see LICENSE.