/Predictive-Maintainence

Forecasting machine failure using a time series approach.

Primary LanguageR

Predictive-Maintainence

  1. Created a univariate Health Index as a function of many sensors using regression.
  2. Modelled these univariate time series using AutoRegressive Integrated Moving Average (ARIMA) and Shallow Neural Network.

This repository is the code base for this article

Preprocessing.R file is used to preprocess the data and generate 100 univariate time series for each of the engines.

Modelling.R file contains the ARIMA and MLP models along with the forecast plots.