Machine Failure Prediction

This project involves creating a machine learning model to predict machine failure based on various features. The project is divided into several parts, including data visualization, model selection, model training, and creating a Tkinter-based application for real-time predictions.

Project Overview

1. Data Visualization and Feature Selection: Analyzes the dataset to visualize relationships and select the most relevant features.
2. Model Selection with GridSearchCV: Determines the best model and hyperparameters using GridSearchCV, from Kaggle.
3. Model Training: Trains a Gradient Boosting model with the selected features and parameters.
4. Tkinter Application: Provides a graphical user interface (GUI) for real-time feature adjustments and predictions.

Requirements

• Python 3.10.14
• Libraries: requirements.txt

Install the required libraries using pip:

pip install -r requirements.txt

Data

Overview

This dataset contains sensor data collected from various machines, intending to predict machine failures in advance. It includes a variety of sensor readings as well as recorded machine failures. Dataset has 944 row (1 repeated row.) and has no NaN values, from Kaggle

Features

• footfall: The number of people or objects passing by the machine.
• tempMode: The temperature mode or setting of the machine.
• AQ: Air quality index near the machine.
• USS: Ultrasonic sensor data, indicating proximity measurements.
• CS: Current sensor readings, indicating the electrical current usage of the machine.
• VOC: The level of volatile organic compounds detected near the machine.
• RP: The machine parts' rotational position or RPM (revolutions per minute).
• IP: Input pressure to the machine.
• Temperature: The operating temperature of the machine.
• fail: Binary indicator of machine failure (1 for failure, 0 for no failure).

Target Distribution

GUI

Results

Results with Data

• Features: footfall tempMode AQ USS CS VOC RP IP Temperature fail
• Test Size: 0.2
• Data Size: 944
• Results:

   Fitting 5 folds for each of 16 candidates, totalling 80 fits
   Best parameters for RandomForest: {'max_depth': None, 'n_estimators': 10}
   Accuracy for RandomForest: 0.8677248677248677
   Fitting 5 folds for each of 20 candidates, totalling 100 fits
   Best parameters for GradientBoosting: {'learning_rate': 0.1, 'n_estimators': 50}
   Accuracy for GradientBoosting: 0.8888888888888888
   Fitting 5 folds for each of 8 candidates, totalling 40 fits
   Best parameters for LogisticRegression: {'C': 0.01}
   Accuracy for LogisticRegression: 0.8783068783068783
   Fitting 5 folds for each of 16 candidates, totalling 80 fits
   Best parameters for SVM: {'C': 0.05, 'kernel': 'linear'}
   Accuracy for SVM: 0.873015873015873

   Best model: GradientBoosting with accuracy: 0.8888888888888888 

Feature Distribution

Correlation Matrix

Results with Feature Reduced Data

• Features: AQ USS VOC RP IP Temperature fail
• Test Size: 0.2
• Data Size: 944
• Results:

   Fitting 5 folds for each of 16 candidates, totalling 80 fits
   Best parameters for RandomForest: {'max_depth': None, 'n_estimators': 50}
   Accuracy for RandomForest: 0.873015873015873
   Fitting 5 folds for each of 20 candidates, totalling 100 fits
   Best parameters for GradientBoosting: {'learning_rate': 0.05, 'n_estimators': 100}
   Accuracy for GradientBoosting: 0.8888888888888888
   Fitting 5 folds for each of 8 candidates, totalling 40 fits
   Best parameters for LogisticRegression: {'C': 0.01}
   Accuracy for LogisticRegression: 0.8783068783068783
   Fitting 5 folds for each of 16 candidates, totalling 80 fits
   Best parameters for SVM: {'C': 0.05, 'kernel': 'linear'}
   Accuracy for SVM: 0.873015873015873

   Best model: GradientBoosting with accuracy: 0.8888888888888888

Feature Distribution

Correlation Matrix