kth.sePublications KTH
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf
From Raw Data to Cognitive Maps: A Design Science Approach to Visualizing Complex Automotive Software System
KTH, School of Electrical Engineering and Computer Science (EECS).
2026 (English)Independent thesis Advanced level (degree of Master (Two Years)), 20 credits / 30 HE creditsStudent thesisAlternative title
Från rådata till kognitiva kartor : En designvetenskaplig metod för att visualisera komplexa fordonsprogramvarusystem (Swedish)
Abstract [en]

Modern automotive industry have evolved into highly complex networks of software components, electronic control units (ECUs), a never-ending number of signals, and properties. Understanding these multi-format, large-scale, highly-nested software structures is a significant cognitive challenge, even for domain experts. This thesis presents an interactive data visualization artifact developed following the Design Science Research Methodology (DSRM) and making use of knowledge graphs. The artifact processes heterogeneous vehicle inputs, including ARXML, YAML, and binary files∗. A unified Python pipeline, together with ReactFlow, is then used to transform the raw, disconnected data into comprehensive network models, enabling developers to build clear mental models of complex software structures. The artifact is evaluated through a comprehensive usability study involving 13 domain experts from Volvo Cars across 9 technical tasks, measuring completion times, interaction frequency, and usability metrics. The evaluation shows that the proposed visualization interface improves task efficiency by over 80%, greatly reducing the average task completion time. Additionally, interface interaction costs were significantly reduced, with the average number of clicks per task decreasing by over 60%. Furthermore, targeted cognitive tasks involving the tracking of sequential data flows showed drastically improvements, with the mean task completion time for identifying dependent and referenced modules decreasing from over 100 seconds to just 3.16 seconds. Lastly, post-test evaluation showed an “excellent” average usability score on the SUS scale. Overall, this study indicates that the knowledge graph-based visualization enhances usability, improves understanding and helps developers build clearer mental models of complex automotive data. However, future studies should consider expanding the evaluation to include broader cognitive and behavioral measures, such as memory recall and eye-tracking analysis.

Abstract [sv]

Den moderna bilindustrin har utvecklats till mycket komplexa nätverk av programvarukomponenter, elektroniska styrenheter (ECU) samt tusentals signaler och egenskaper. Att förstå dessa storskaliga och djupt kapslade strukturer är därför en betydande kognitiv utmaning för experter inom området. För att adressera detta problem presenterar denna studie ett interaktivt visualiseringsverktyg baserat på Design Science Research Methodology och kunskapsgrafer. I likhet med en digital karta i en komplex stad fungerar verktyget som en karta över komplexa dataflöden. Det integrerar spridda tekniska format – såsom ARXML, YAML och binära loggfiler – till en enhetlig visuell nätverkskarta, vilket stödjer utvecklares förståelse och spårning av systemrelationer. Det designade webbgränssnittet utvärderas av 13 ingenjörer från Volvo Cars, där varje deltagare genomförde 9 typiska uppgifter. Forskare mätte slutförandetid, antal musklick samt användbarhet via System Usability Scale (SUS). Resultaten visar tydliga förbättringar. Det nya gränssnittet ökade uppgiftseffektiviteten med 80,3 Särskilt stora förbättringar observerades i dataspårningsuppgifter, där tiden för att hitta beroende moduler minskade från 106,5 sekunder till 3,16 sekunder. Systemet fick även ett SUS-värde på 83,5, vilket motsvarar “Utmärkt” användbarhet. Resultaten visar att kunskapsgraf-baserad visualisering kan minska komplexitet, förbättra förståelse och öka effektivitet. Framtida arbete kommer att undersöka kognitiva effekter djupare med hjälp av ögonspårning och minnestester.

Place, publisher, year, edition, pages
2026. , p. 54
Series
TRITA-EECS-EX ; 2026:407
Keywords [en]
Software architecture (Software organization and properties), Automotive software architecture (Embedded systems; software architecture), Design science research (Methodology; artifact design; evaluation), Data flow visualization (Information visualization; data representation), Interactive visualization (Human-computer interaction; visualization), Web-based user interfaces (Web applications; user interface design)
Keywords [sv]
Mjukvaruarkitektur (Mjukvaruorganisation och egenskaper), Automotive mjukvaruarkitektur (Inbyggda system; mjukvaruarkitektur), Design science- forskning (Metodologi; artefaktdesign; utvärdering), Dataflödesvisualise- ring (Informationsvisualisering; datarepresentation), Interaktiv visualisering (Människa–datorinteraktion; visualisering), Webbaserade användargränssnitt (Webbapplikationer; användargränssnittsdesign)
National Category
Computer and Information Sciences
Identifiers
URN: urn:nbn:se:kth:diva-391280OAI: oai:DiVA.org:kth-391280DiVA, id: diva2:2112329
External cooperation
Volvo Cars
Educational program
Master of Science - Interactive Media Technology
Supervisors
Examiners
Available from: 2026-11-03 Created: 2026-10-22 Last updated: 2026-11-03Bibliographically approved

Open Access in DiVA

fulltext(26505 kB)24 downloads
File information
File name FULLTEXT01.pdfFile size 26505 kBChecksum SHA-512
b95e0620992f24b8a85f3b3506f520d9975c86d0fa234c8761ca1240d8f231fe3b1c26da42dda90434503077ff68765a3834cb8869d6bed0360398924f9d84cc
Type fulltextMimetype application/pdf

By organisation
School of Electrical Engineering and Computer Science (EECS)
Computer and Information Sciences

Search outside of DiVA

GoogleGoogle Scholar
Total: 24 downloads
The number of downloads is the sum of all downloads of full texts. It may include eg previous versions that are now no longer available

urn-nbn

Altmetric score

urn-nbn
Total: 848 hits
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf