Deakin University
Browse

File(s) under permanent embargo

IoT-based system health management infrastructure as a service

conference contribution
posted on 2018-10-29, 00:00 authored by Gokul Thirunavukkarasu, Benjamin ChampionBenjamin Champion, Ben HoranBen Horan, Mohammadmehdi Seyedmahmoudian, A Stojcevski
Customization, enhanced quality of streamlined maintenance services and uplifted productivity are some of the key highlights from the rapidly evolving concept of Industry 4.0. IoT (Internet of things) based service infrastructure models designed for delivering enterprise services with capabilities of pro-actively sensing malfunctions and responding with preventive measures to streamline the automated service offered is one of the prime application of this concept. Continuous maintenance services increase the optimum through-life cost and in-service life cycle of the product providing the customer with the feel of full ownership. In-service feedbacks also help the manufactures to identify issues with respect to the designs and improve it in the future versions. In this paper, as a proof of concept a cloud-based IoT service infrastructure for providing real-time prognostic and supervised vehicle maintenance system is proposed. This proposed system aims at providing an enterprise service infrastructure to the registered vehicle service centers to keep track of the real-time vehicle diagnostic information of their client's vehicle over cloud and use prognostic algorithms to identify any malfunctions or abnormal behavior of the vehicles for automatically scheduling a service appointment and automating the maintenance cycle of the vehicle. In addition to this, the system provides features like remote supervision and diagnostics maintenance enabling technicians to fix issues remotely, ensuring streamlined and reliable service. Initially, before building the proposed prototype system, a few experimental trails where conducted for analyzing the use of different IoT models used in the development to identify the best-suited approach. The results indicated that the publisher-subscriber (NodeJS) based model outperforms the request-response (PHP) based model in terms of the hits per second and mean request time for an increased number of active users. The results of the initial tests justify the reason for the using the publisher-subscriber based IOT architecture. The conceptualized enterprise infrastructure illustrated in the manuscript aims at providing a streamlined maintenance service.

History

Event

IEEE Communications Society. Conference (2018 : Singapore)

Series

IEEE Communications Society Conference

Pagination

55 - 61

Publisher

Association for Computing Machinery

Location

Singapore

Place of publication

New York, N.Y.

Start date

2018-10-29

End date

2018-10-31

ISBN-13

9781450365765

Language

eng

Publication classification

E Conference publication; E1 Full written paper - refereed

Copyright notice

2018, Association for Computing Machinery

Editor/Contributor(s)

[Unknown]

Title of proceedings

CCIOT 2018 : Proceedings of the 2018 International Conference on Cloud Computing and Internet of Things