/GE-Case-Study

This case study focuses on predicting aircraft remaining useful life (rul), by analyzing data from different datasets given by GE Aerospace.

Primary LanguageHTML

File Table of Contents (In Order of Project Process):

GE_CaseStudy_Markdown.html
GE_Cleaned_Data_Corrected.csv
GE_CaseStudy_Profile.ipynb
GE_Tableau.twbx
GE_Model_groupby.html
Model_Data.csv
GE_Predictive_Model_And_App.ipynb
final_model.pki
GE_Pres_Final.qmb

GE_CaseStudy_Markdown

(html located in GE_CaseStudy_Github.io)


  • RMarkdown of the merged dataset used for starting the case study.

GE_Cleaned_Data_Corrected

  • Technically corrected, tidy, and consistency tested holistic dataset.

GE_CaseStudy_Profile

  • Jupyter notebook using Pandas_Profiling to create a series of automated statistics.

GE_Tableau

  • Tableau interactive dashboard to showcase relevant features and interesting results.

GE_Model_groupby

(html located in GE_CaseStudy_Github.io)


  • RMarkdown indicating which aggregated variables was used for the predictive model.

Model_Data

  • A dataset of the variables statistically significant to predicting aircraft remaining useful life.

GE_Predictive_Model_And_App

  • Jupyter notebook schowcasing the process of building, training, and evaluating the 'best fit' predictive model for the variable 'rul'. The model is then created into an embedded app.

final_model

  • A pickle file of the machine learning model for easy storing.

GE_Pres_Final

(html located in GE_CaseStudy_Github.io)


  • Quarto presentation highlighting the process of the case study.