Backorder-prediction- Intership-Project-1

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).

Data Analysis

In the Train dataset we are provided with 23 columns(Features) of data.

Sku(Stock Keeping unit) : The product id — Unique for each row so can be ignored
National_inv : The present inventory level of the product
Lead_time : Transit time of the product
In_transit_qty : The amount of product in transit
Forecast_3_month , Forecast_6_month , Forecast_9_month : Forecast of the sales of the product for coming 3 , 6 and 9 months respectively
Sales_1_month , sales_3_month ,sales_6_month , sales_9_month : Actual sales of the product in last 1 , 3 ,6 and 9 months respectively
Min_bank : Minimum amount of stock recommended
Potential_issue : Any problem identified in the product/part
Pieces_past_due: Amount of parts of the product overdue if any
Perf_6_month_avg , perf_12_month_avg : Product performance over past 6 and 12 months respectively
Local_bo_qty : Amount of stock overdue
Deck_risk , oe_constraint, ppap_risk, stop_auto_buy, rev_stop : Different Flags (Yes or No) set for the product
Went_on_backorder : Target variable

Out of the 23 features given in the dataset 15 are numerical and 8(including the target variable) are categorical features.

Approach

The main goal is to predict the whether a product comes in backorder or not based on different factors available in the dataset.

Data Exploration : Exploring dataset using pandas,numpy,matplotlib and seaborn.
Data visualization : Ploted graphs to get insights about dependend and independed variables.
Feature Engineering : Removed missing values and created new features as per insights.
Model Selection I : Tested all base models to check the base accuracy. Also ploted and calculate Performance Metrics to check whether a model is a good fit or not.
Model Selection II : Performed Hyperparameter tuning using RandomsearchCV.
Pickle File : Selected model as per best accuracy and created pickle file using pickle library.
Webpage & deployment : Created a webform that takes all the necessary inputs from user and shows output. After that I have deployed project on AWS .

Technologies Used

Pycharm Is Used For IDE.
For Visualization Of The Plots Matplotlib , Seaborn Are Used.
AWS is Used For Model Deployment.
Cassandra Database Is Used To As Data Base.
Front End Deployment Is Done Using HTML , CSS.
Flask is for creating the application server and pages.
Git Hub Is Used As A Version Control System.
josn is for data validation processes.
os is used for creating and deleting folders.
csv is used for creating .csv format file.
numpy is for arrays computations and mathematical operations
pandas is for Manipulation and wrangling structured data
scikit-learn is used for machine learning tool kit
pickle is used for saving model
XgBoost is used for XgBoostClassifier Implementation.
Nearmiss Imbalance is used for handling Imbalance Data.

User InterFace

Home Page

Predict Page

Back Order Prediction Project Video

Back.Order.Prediction.mp4

Deployment Links

AWS link : http://ec2-18-133-246-122.eu-west-2.compute.amazonaws.com:8080

Develoyment Process Video

Development_Process_Using_AWS.mp4

Run Locally

Clone the project

  git clone https://github.com/Vikas2201/Backorder-prediction-1

Go to the project directory

  cd Backorder-prediction-1

Install dependencies

  pip install -r requirements.txt

Setting up the Controllers files

    Update the values inside the Controllers folder

Run the app.py

  python app.py

Usage

In Development If You Want to contribute? Great!

To fix a bug or enhance an existing module, follow these steps:

Fork the repo
Create a new branch

   git checkout -b new-feature

Make the appropriate changes in the file
Commit your changes

     git commit -am "New feature added"

Push to the branch

      git push origin new-feature

Create a pull request

Directory Tree

└── Backorder-Prediction
    ├── application_logging
    │   └── logging.py        
    ├── Controllers
    │   └── DBconnection_info.yaml
    ├── Data_Information 
    │   └── Null_Values.csv
    ├── data_ingestion
    │   ├── __pycache__
    │   └── data_loader.py
    ├── data_preprocessing
    │   ├── __pycache__
    │   └── preprocessing.py
    ├── EDA 
    │   ├── EDA-part1.ipynb
    │   └── EDA-part2.ipynb
    ├── model 
    │   └── Random Forest
    │       └── Random Forest.sav 
    ├── model_finder
    │   ├── __pycache__
    │   └── Model.py
    ├── Prediction_logs 
    │   └── predictionlog.txt 
    ├── Prediction_Output_File 
    │   └── Predictions.csv 
    ├── ReplaceMissingwithNull
    │   ├── __pycache__
    │   ├── __init__.py
    │   └── transformer.py
    ├── Save_Models
    │   ├── __pycache__
    │   └── save_methods.py
    ├── static 
    │   ├── backorder.jpg
    │   ├── Style1.css
    │   └── Style2.css 
    ├── Templates
    │   ├── index.html 
    │   └── predict.html 
    ├── Tools
    │   ├── __pycache__
    │   ├── __init__.py
    │   ├── DBconnector.py
    │   ├── training_logFilescreater.py
    │   └── YamlParser.py
    ├── Traning_batch_Files
    │   └── BackOrder_08012020_120000.csv
    ├── Training_Database_operations 
    │   ├── __init__.py
    │   ├── __pycache__
    │   └── Database_handler.py
    ├── Traininig_dataValidation 
    │   ├── __pycache__
    │   └── RawtrainingValidation.py 
    ├── Training_FilesfromDB
    │   └── InputFile.csv 
    ├── Training_Logs
    │   ├── columnValidationLog.txt
    │   ├── DataBaseConnectionLog.txt
    │   ├── Datapreprocessing_logs.txt
    │   ├── GeneralLog.txt
    │   ├── TmissingValuesInColumn.txt
    │   ├── ModelTrainingLog.txt
    │   ├── nameValidationLog.txt
    │   ├── TrainingDatabseInfo.txt
    │   ├── TrainingMainLog.txt
    │   ├── valuesfromSchemaValidationLog.txt
    │   └── yaml_parser.txt
    ├── Training_Raw_Files_Validated
    │   └── BackOrder_08012020_120000.csv
    ├── training_val_linkage.py
    ├── training_file.py
    ├── app.py
    ├── prediction.py
    ├── requierements.txt 
    ├── Training_Schema.json
    └── secure-connect-backorder-prediction.zip

Conclusions

We Developed A Product Backorder Predictive Model With The Capability Of Identifying Items To Be Backordered Using Machine Learning Models. The Proposed Approach Accept Input Data Was Pre-Processed By Way Of Missing Values Imputation, Non-Numeric To Numeric Feature Conversion And Normalization, And Split Into Training And Test Set. The Training Set Is Passed Into A Data Balancing Module To Ensure Equal Class Distribution And Avoid Biasness In Learning Model Decisions. The Imbalanced Training Data Were Subjected Sampling As We Concurrently Fed The Data Into Sampling Techniques Fed Into ML Models To Predict Product Backorders. The Predictive Models Were Validated On Test Data And Their Performances Were Evaluated. The Evaluation Of The Result Obtained Showed By Precision, Recall , AUC Score , Accuracy And F1-Score.