Modelling industrial technical compromises in production systems with causal loop diagrams

Quang Huan Dong, Birgit Vogel-Heuser

Research output: Contribution to journalConference articlepeer-review

5 Scopus citations

Abstract

Automated production systems (aPS) have a long lifetime, which requires significant maintenance cost. During the commissioning phase of aPS, which often occurs at customer sites, the on-site engineers are always under high pressure because the customers would like to start the production at the earliest. There, some compromises, so-called technical debt (TD), might be chosen. A technical compromise taken in the commissioning phase can provide a short-term benefit, but could yield a long-term negative impact on the maintenance phase. This study aims to apply Causal Loop Diagram (CLD) to model technical compromises in the aPS domain. A concept, namely CLD+TD, is proposed. Application examples of CLD+TD on some technical compromises collected from an industrial company working in automation are reported. A loop dominance analysis on a CLD+TD model is performed to derive the circular trend of technical compromise over time; thus, relevant maintenance activities could be planned accordingly. In addition, challenges leading to the technical compromises addressed in the presented use cases can support practitioners in developing appropriate solutions to manage the reported technical compromises in the commissioning phase.

Original languageEnglish
Pages (from-to)212-219
Number of pages8
JournalIFAC Proceedings Volumes (IFAC-PapersOnline)
Volume54
Issue number4
DOIs
StatePublished - 2021
Event4th IFAC Conference on Embedded Systems, Computational Intelligence and Telematics in Control CESCIT 2021 - Valenciennes, France
Duration: 5 Jul 20217 Jul 2021

Keywords

  • Case study
  • Causal loop diagram
  • Production systems
  • Technical debt

Fingerprint

Dive into the research topics of 'Modelling industrial technical compromises in production systems with causal loop diagrams'. Together they form a unique fingerprint.

Cite this