Skip to content

Latest commit

 

History

History
76 lines (56 loc) · 4.11 KB

README.md

File metadata and controls

76 lines (56 loc) · 4.11 KB
Raw

BACK ORDER PREDICTION

Problem Statement

  • Backorders are unavoidable, but by anticipating which things will be backordered,planning can be streamlined at several levels, preventing unexpected strain on production, logistics, and transportation.
  • ERP systems generate a lot of data (mainly structured) and also contain a lot of historical data; if this data can be properly utilized, a predictive model to forecast backorders and plan accordingly can be constructed.
  • Based on past data from inventories, supply chain, and sales, classify the products as going into backorder (Yes or No).
  • The Problem is to prevent the unexpected strain on production,logistcs and transportation.
  • From then problem statement we need reduce the both False Positve and False Negative.
  • False Negative reduction will help us to prevent the unexpected strain on production,logistics and transportation
  • False positive is not that much important to the company because even if it is predicted as product is going to be backordered but actually is not backorded.

Project Architecture

Project Architecture

Project PipeLine Stages

In this project, I have developed a machine learning model that predicts the likelihood of back orders for products. The project pipeline is as follows:

Data Ingestion:

  • The project begins with data ingestion from S3 buckets. The data is retrieved and preprocessed for further analysis.

Data Validation:

  • The data is carefully validated to ensure its integrity and consistency. This step involves checking for missing values, handling outliers, and verifying data types.

Data Transformation:

  • To prepare the data for modeling, various transformation techniques are applied, such as feature scaling, one-hot encoding, or label encoding. This step ensures that the data is in a suitable format for model training.

Model Training:

  • Machine learning algorithms are utilized to train a predictive model using the transformed data. Several models, such as Random Forest, Gradient Boosting, or Support Vector Machines, can be explored to identify the best-performing model.

Model Evaluation:

  • The trained model is evaluated using appropriate evaluation metrics like accuracy, precision, recall, or F1 score. This step helps assess the model's performance and identify areas for improvement.

Model Pusher:

  • Once a satisfactory model is obtained, it is pushed to a production environment or deployed as an API to make predictions on new data.

Frontend Website:

  • To provide a user-friendly interface for interaction, I have used Streamlit to create a frontend website. Users can input relevant information and receive predictions for back orders based on the trained model.
  • Deployed stremlit app link
  • Click on link Backorder prediction

Project PipeLine Stages

Usage

To run the project locally, follow these steps:

  1. Clone the repository:
git clone [https://github.com/your-username/backorder-prediction.git](https://github.com/medashabari/BackorderPrediction.git)
  1. Install the required dependencies:
pip install -r requirements.txt
  1. Run the Streamlit application:
streamlit run main.py
  1. Access the application in your browser at http://localhost:8501.

Contributing

  • Contributions are welcome! If you want to contribute to this project, please follow these steps:
  1. Fork the repository.
  2. Create a new branch for your feature or bug fix.
  3. Make the necessary modifications.
  4. Commit your changes and push the branch.
  5. Submit a pull request with a detailed description of your changes.

Contact

If you have any questions or would like to collaborate on this project, please feel free to reach out to me. You can contact me via email at [email protected] or connect with me on LinkedIn.

Let's make accurate backorder predictions together!