Skip to content

Latest commit

 

History

History
76 lines (55 loc) · 2.7 KB

README.md

File metadata and controls

76 lines (55 loc) · 2.7 KB

Sprout

Sprout is a simple Business Intelligence (BI) system designed for retail stores. It consists of six separate but linked Dash applications running on different ports. The system provides key insights into inventory, sales/performance, customer behaviour, business goals, advanced analytics, and reports with customizable queries.

Features

Sprout consists of six dashboards:

  1. Inventory Dashboard - Displays statistics on current inventory levels assiting in inventory management.
  2. Sales Dashboard - Shows sales performance and key metrics to track business growth.
  3. Customer Dashboard - Analyzes customer behavior and statistics.
  4. Goals Dashboard - Tracks existing business goals and allows users to set new ones.
  5. Analysis Dashboard - Provides advanced analytical insights, including:
    • Time Series Sales Prediction using Facebook's Prophet model.
    • Customer Lifetime Value (CLV) Analysis.
    • Product Affinity Analysis.
  6. Queries Section - A collection of reports with advanced filtering mechanisms (not a dashboard but an interactive reporting tool).

Installation and Running the Project

Prerequisites

  • Python installed on your system
  • Required dependencies (see requirements.txt)

Steps to Run the Application

  1. Clone the repository: 
    git clone <repository_url>
  2. Navigate into the project directory: 
    cd Sprout
  3. Navigate to the dashboards folder: 
    cd dashboards
  4. Run all the applications using the provided PowerShell script: 
    ./run_all_apps.ps1
  5. Open a web browser and visit one of the running dashboards, e.g., http://localhost:8050.

Notes

  • This project originally included a separate home page for access control, but it is not included in this version.
  • Each dashboard runs on a separate port and contains navigation links to the other dashboards.
  • Use the side menu to explore different dashboards.

Technologies Used

  • Dash (for building web-based dashboards)
  • Plotly (for interactive visualizations)
  • Prophet (for time series forecasting)
  • Pandas & NumPy (for data manipulation)
  • Statsmodels (for statistical analysis)
  • Dash Mantine Components, Dash Bootstrap Components (for UI enhancements)

Future Improvements

  • User management and Access Control.
  • Enhancing visualizations and with more interactivity.
  • Expanding analytics with additional/better predictive models.
  • Use of machine learning models for clv and products affinity analysis instead of mathematical/statistical methods.

License

This project is open-source. Feel free to modify and extend it as needed.

For questions or contributions, feel free to reach out!