DataCo Smart Supply Chain Analytics Dashboard

🚚🏭🏗️ Welcome to the DataCo SMART SUPPLY CHAIN Analysis Dashboard! This interactive Streamlit application provides insights into the DataCo Smart Supply Chain dataset, focusing on various aspects of supply chain performance.

Table of Contents

• Project Overview
• Features
• Navigation
• Installation
• Usage

Project Overview

This project explores the DataCo Smart Supply Chain dataset to gain insights into the supply chain performance, focusing on key metrics such as sales, profit, and delivery time.

Key Areas of Focus:

• IoT and Data Growth: The widespread adoption of IoT has generated vast amounts of data, valuable for uncovering insights and enhancing decision-making.

• Customer Segmentation: This project utilizes the DataCo dataset to conduct customer segmentation, enabling better customer understanding and revenue growth.

• Model Selection Challenge: With multiple data analysis methods and models available, choosing the right one is crucial as model performance varies with data parameters.

• Comparative Studies: Prior studies (e.g., Carbonneau, Hill, Vakili, Ahmed) have compared traditional forecasting methods with neural networks and machine learning models, revealing varying levels of performance across techniques.

• Project Objective: This study plans to compare 9 machine learning classifiers and 7 regression models against neural networks. Key tasks include fraud detection, late delivery prediction, sales forecasting, and demand prediction.

Model Performance (Subject to Change):

• Classification models: Logistic Regression, SVM, k-NN, Random Forest, etc., evaluated for accuracy, recall, and F1 score.
• Regression models: Lasso, Ridge, Random Forest, and others, assessed with MAE and RMSE for sales and demand prediction.

Features

• Data Cleaning and Analysis (IDA): Explore and clean the dataset to prepare it for analysis.
• Exploratory Data Analysis (EDA): Visualize key trends and patterns in the data.
• Missingness Analysis: Understand the distribution of missing values in the dataset.

Navigation

Use the sidebar to navigate between different sections of the app:

Professional of Business

• Overview: A summary of the project and its objectives.
• User Guide, Documentation & References: Detailed project documentation and references.
• Customer Segmentation Analysis: Insights into customer grouping and segmentation.
• Interactable EDA (β-version): Interactive exploratory data analysis.
• Modelling Results: Results and evaluations of predictive models.

Professional of Technology

• IDA: Initial Data Analysis and data cleaning.
• EDA: Exploratory Data Analysis with visualizations.
• Missingness Analysis: Insights into missing data patterns and handling.
• Modelling Metrics: Detailed metrics and evaluation of models.

Usage

• Here you can find the Official App Link [Final Link for End Term] - https://cmse830project-aj-finals.streamlit.app/
• Here you can find the Prototype App Link [Final Link for Mid Term] - https://cmse830project-aj.streamlit.app/

Once the app is running, you can interact with it through your web browser. Navigate between sections using the sidebar to explore the data insights and visualizations.

Feel free to contribute or reach out for any questions!