Skip to main content

Lund University Publications

LUND UNIVERSITY LIBRARIES

Evaluating TRIZ with and without LLM support : an experimental study on engineering problem-solving

Čok, Vanja ; Motte, Damien LU orcid ; Hmina, Khadija ; El Hassani, Ibtissam LU ; Demšar, Ivan ; Tavčar, Jože LU orcid and Vukašinović, Nikola (2026) 19th International Design Conference, DESIGN 2026 In Proceedings of the Design Society 6. p.507-516
Abstract
This paper examines integrating Large Language Models (LLMs) into the TRIZ contradiction matrix (TRIZ-C+LLM) to support engineering students in creative problem-solving. Experiments with three problems show that LLMs did not always improve design quality for complex tasks but reduced cognitive workload, improved understanding of contradictions, and increased perceived usefulness. Prompting experience strongly influenced outcomes, highlighting both the promise and limits of combining TRIZ with generative AI.
Please use this url to cite or link to this publication:
author
; ; ; ; ; and
organization
publishing date
type
Chapter in Book/Report/Conference proceeding
publication status
published
subject
keywords
TRIZ, Large language models, LLM, Generative AI
host publication
Proceedings of the Design Society. International Design Conference, DESIGN 2026 18-21 May 2026, Dubrovnik, Croatia
series title
Proceedings of the Design Society
volume
6
pages
507 - 516
publisher
Cambridge University Press
conference name
19th International Design Conference, DESIGN 2026
conference location
Dubrovnik, Croatia
conference dates
2026-05-18 - 2026-05-21
ISSN
2732-527X
DOI
10.1017/pds.2026.10409
language
English
LU publication?
yes
id
80ee1fac-a63c-4753-a9ad-952675c2433e
date added to LUP
2026-07-06 11:15:48
date last changed
2026-08-26 11:52:21
@inproceedings{80ee1fac-a63c-4753-a9ad-952675c2433e,
  abstract     = {{This paper examines integrating Large Language Models (LLMs) into the TRIZ contradiction matrix (TRIZ-C+LLM) to support engineering students in creative problem-solving. Experiments with three problems show that LLMs did not always improve design quality for complex tasks but reduced cognitive workload, improved understanding of contradictions, and increased perceived usefulness. Prompting experience strongly influenced outcomes, highlighting both the promise and limits of combining TRIZ with generative AI.}},
  author       = {{Čok, Vanja and Motte, Damien and Hmina, Khadija and El Hassani, Ibtissam and Demšar, Ivan and Tavčar, Jože and Vukašinović, Nikola}},
  booktitle    = {{Proceedings of the Design Society. International Design Conference, DESIGN 2026 18-21 May 2026, Dubrovnik, Croatia}},
  issn         = {{2732-527X}},
  keywords     = {{TRIZ; Large language models; LLM; Generative AI}},
  language     = {{eng}},
  month        = {{07}},
  pages        = {{507--516}},
  publisher    = {{Cambridge University Press}},
  series       = {{Proceedings of the Design Society}},
  title        = {{Evaluating TRIZ with and without LLM support : an experimental study on engineering problem-solving}},
  url          = {{https://lup.lub.lu.se/search/files/254818725/Cok_al_Design2026.pdf}},
  doi          = {{10.1017/pds.2026.10409}},
  volume       = {{6}},
  year         = {{2026}},
}