Data-driven Maintenance Service (DDMS)

Context

The quick advance in the last years of fields like Industrial Internet of Things (IIoT), Cloud Computing and Big Data allows manufacturing companies to capture more and better data from their machines, in a way that was simply impossible to do just a few years ago. These technologies are the fulcrum to allow companies to move from a paper-based, mostly manual maintenance approach to a new, more automated, data-based strategy.

Idea

But to get to that point, much work remains to be done, to create easy to implement, affordable solutions. This project will tackle this problem, with the main objective to help manufacturing companies to improve their asset maintenance strategy, by exploiting the potential of data analysis of machine data in maintenance, in two different approaches:

 

  1. Reactive maintenance: to implement a data-driven Helpdesk service that will assist technical assistants when dealing with machine failure and anomalous situations.
  2. Condition-based and Predictive maintenance: to implement a data-driven notification service that can predict future likely issues.

Aim

Technical Goal 1: Advanced Helpdesk Service

 

A data-driven application that will guide technicians by using machine monitored data to help understand machine issues. This service will consist in a service oriented to reactive maintenance, where data preprocessing and analysis will be used to help diagnose machine issues, and condition-based and predictive maintenance service that will evaluate the probability of future issues and detect anomalous situations.

 

Technical Goal 2: IoT and Big Data will be used to develop a flexible platform to provide data-driven solutions to solve common maintenance challenges, with a special focus in predictive maintenance, for a wide range of machines

Challenge

While the use of data science and IoT for predictive maintenance is a hot topic, it has not been yet widely implemented. Some of the reasons are the following:

 

  • Many companies are not yet fully aware of the potential of their machine data. Some of them are reluctant to spend resources in having their data analyzed.
  • Some of the data required to perform a suitable analysis can be difficult to obtain.

Real-time Integration of Machine and Sensor Data for Predictive Maintenance

PROJECT MANAGER

Carlos Garcia

Senior Technologist

Technical Lead

Anthony Faustine

Senior Industrial Analytics Researcher

THEMATIC PILLAR

Digitisation

FOLLOW US

A project funded under the SMART EUREKA CLUSTER on Advanced Manufacturing programme through Enterprise Ireland – Grant IR20190049