A messaging library for distributed modeling

Research output: Contribution to journalConference articlepeer-review

Abstract

The ongoing digital transformation of industry is most clearly reflected in the increasing collection and analysis of data from various sources. Among these are sensor equipped machinery, telemetry in logistics or audiovisually monitored production floors. In order to utilize the data, e. g., to predict machinery malfunctions, its technically smooth consolidation is crucial. Therefore, numerous data interchange formats, protocols and middleware emerged over the past years. However, until today there is no gold standard technology stack for industrial data analysis, for multiple reasons, including the applications' heterogeneity. In this work, we present a software library which aims at decoupling messaging protocols and patterns from their implementation to overcome incompatibilities and, thus, facilitate data consolidation for software engineers. Moreover, we show how to use the library for rapidly modeling distributed cyber-physical systems using an integrated schema generation mechanism. Based on one real-world and one synthetic use case, we evaluate the library's applicability, discuss open issues and outline planned features.

Original languageEnglish
Pages (from-to)606-615
Number of pages10
JournalProcedia Computer Science
Volume232
DOIs
Publication statusPublished - 2024
Event5th International Conference on Industry 4.0 and Smart Manufacturing, ISM 2023 - Lisbon, Portugal
Duration: 22 Nov 202324 Nov 2023

Keywords

  • Industrial Machine Learning
  • Industry 4.0
  • Message-Oriented Middleware
  • Open Source
  • Software Design

Fingerprint

Dive into the research topics of 'A messaging library for distributed modeling'. Together they form a unique fingerprint.

Cite this