Skip to content

Big Data In & OUT services for Industry 4.0: from shopfloor to post-sales

Notifications You must be signed in to change notification settings

softlab-unimore/SBDIOI40

Repository files navigation

Big Data In & OUT services for Industry 4.0: from shopfloor to post-sales

SBDIO I4.0 is an industrial research project which aims to sustain local industries in their digitalization process, fostering transitions from a product economy to a services economy.

SBDIO I4.0 project stems from the needs of all the companies belonging to the industrial automation chain and based in the Emilia-Romagna Region, with the goal to simplify the transition from a product economy to a services economy.

The main objective of the project is the creation of a smart platform capable to offer a double servitization process for the industries: in the production (shopfloor) and in the post-sales services. This progressive transition can be developed by using an innovative approach based on Artificial Intelligence and Machine Learning algorithms applied on big data.

The project involves 5 industrial research laboratories belonging to 3 Universities and 1 research center in addition to the participation of 7 companies.

Goal

  • Big Data: to adopt new big data technology in industrial field
  • Industry 4.0: transition from a product economy to a services economy
  • Artificial Intelligence: servitization of both production and post-sales services
  • Machine Learning: Predictive models to anticipate maintenance actions and breakdown

Library

Requirements

  • Python: Python 3.*
  • Packages: requirements.txt

Installation

$ cd source

$ virtualenv -p python3 venv

$ source venv/bin/activate

$ pip install -r requirements.txt

How to Use

Inside the timely folder, there are the models and transformed implemented and tested in industrial use case, instead in demo there is an example of anomaly detection application

Anomaly Detection Model

  • CNN AutoEncoder
  • LSTM AutoEncoder
  • MLP AutoEncoder
  • Setup Clustering
  • Isolation Forest
  • One Class SVM
  • Local Outlier Factor (LOF)
  • PCA Anomaly Detection
  • One Threshold

About

Big Data In & OUT services for Industry 4.0: from shopfloor to post-sales

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published