Tomislav Martinec
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Tomislav Martinec has a total of 48 co-authors. Here are the top co-authors by number of shared publications:
- Stanko Škec
Faculty of Mechanical Engineering and Naval Architecture, University of Zagreb, Ivana Lucica 5, 10000 Zagreb, Croatia
13 shared publications - Nikola Horvat
University of Zagreb
5 shared publications - Mario Štorga
University of Zagreb
4 shared publications Fanika Lukačević
University of Zagreb Faculty of Mechanical Engineering and Naval Architecture, Croatia
3 shared publicationsMario Štorga
University of Zagreb, Faculty of Mechanical Engineering and Naval Architecture
3 shared publications- Petar Kosec
Faculty of Mechanical Engineering and Naval Architecture, University of Zagreb, Ivana Lucica 5, 10000 Zagreb, Croatia
3 shared publications
Latest Publications
- Open Access
- 29 May 2026
- Published by MDPI AG in Machine Learning and Knowledge Extraction
- Vol. 8 (6) , 147
Generation of Synthetic Dataset for Part Segmentation Problems
Part segmentation of industrial 3D models is often limited by the lack of sufficiently large and consistently labeled training datasets. This study proposes a workflow for generating synthetic segmentation datasets from robust parametric computer-aided design (CAD) models and evaluates its applicability on a dental abutment case. The workflow includes the definition of a modeling strategy, creation of a robust parametric CAD model, automated generation of valid geometry variants, and preparation of labeled training data for point-cloud-based segmentation. In the experimental part of the study, a synthetic dataset of segmented dental abutment geometries was generated from the developed parametric CAD model and used to train a PointNeXt-S part-segmentation model. The segmentation performance of the trained model was evaluated on manually labeled real-world abutments. Results show that the segmentation of industrial 3D models improved with increasing synthetic training-set size and further improved when data augmentation was applied. The best-performing augmented model achieved a mean Intersection over Union (IoU) of 89.2% on the real-world validation set, compared with 82.4% without augmentation. The findings indicate that parametric-CAD-based synthetic dataset generation can provide an effective basis for training segmentation models for complex industrial geometries.
- 29 May 2026
- Published by MDPI AG in Machine Learning and Knowledge Extraction
- Vol. 8 (6) , 147
Open AccessGeneration of Synthetic Dataset for Part Segmentation Problems
Part segmentation of industrial 3D models is often limited by the lack of sufficiently large and consistently labeled training datasets. This study proposes a workflow for generating synthetic segmentation datasets from robust parametric computer-aided design (CAD) models and evaluates its applicability on a dental abutment case. The workflow includes the definition of a modeling strategy, creation of a robust parametric CAD model, automated generation of valid geometry variants, and preparation of labeled training data for point-cloud-based segmentation. In the experimental part of the study, a synthetic dataset of segmented dental abutment geometries was generated from the developed parametric CAD model and used to train a PointNeXt-S part-segmentation model. The segmentation performance of the trained model was evaluated on manually labeled real-world abutments. Results show that the segmentation of industrial 3D models improved with increasing synthetic training-set size and further improved when data augmentation was applied. The best-performing augmented model achieved a mean Intersection over Union (IoU) of 89.2% on the real-world validation set, compared with 82.4% without augmentation. The findings indicate that parametric-CAD-based synthetic dataset generation can provide an effective basis for training segmentation models for complex industrial geometries.
- Open Access
- 27 February 2026
- Published by MDPI AG in Applied Sciences
- Vol. 16 (5) , 2291
Process Mining in Digital Dental Laboratories: Identifying Iterations Through Actions and Digital Artefacts
Digitalization has reshaped dental laboratory processes through digital tools and artefacts supporting clinician–laboratory collaboration; however, repeated iterations still increase coordination effort and extend delivery times. This study examined how the custom abutment process was executed in a dental laboratory and identified where and why iterations occurred during computer-aided design (CAD) modelling, design verification, and manufacturing preparation. Ten completed orders were selected, and their event log information was analyzed using process mining in Disco, complemented by contextual inquiry with domain practitioners. The analysis reconstructed observed execution from order initiation to delivery and derived a reference representation summarizing the most frequently observed ordering of actions. Across the ten orders analyzed, nine exhibited at least one iteration. Iterations were most frequently observed as returns between CAD modelling and design verification and occurred in four orders, while rescanning occurred in two orders due to insufficient or incompatible initial scan information. Contextual inquiry linked repeated action sequences to changes in digital artefacts and communication exchanges, indicating that iterations were associated with incomplete information or differences in interpretation across roles. The findings show that combining process mining with contextual inquiry enables the identification of iterations and clarifies the conditions under which they emerge.
- 27 February 2026
- Published by MDPI AG in Applied Sciences
- Vol. 16 (5) , 2291
Open AccessProcess Mining in Digital Dental Laboratories: Identifying Iterations Through Actions and Digital Artefacts
Digitalization has reshaped dental laboratory processes through digital tools and artefacts supporting clinician–laboratory collaboration; however, repeated iterations still increase coordination effort and extend delivery times. This study examined how the custom abutment process was executed in a dental laboratory and identified where and why iterations occurred during computer-aided design (CAD) modelling, design verification, and manufacturing preparation. Ten completed orders were selected, and their event log information was analyzed using process mining in Disco, complemented by contextual inquiry with domain practitioners. The analysis reconstructed observed execution from order initiation to delivery and derived a reference representation summarizing the most frequently observed ordering of actions. Across the ten orders analyzed, nine exhibited at least one iteration. Iterations were most frequently observed as returns between CAD modelling and design verification and occurred in four orders, while rescanning occurred in two orders due to insufficient or incompatible initial scan information. Contextual inquiry linked repeated action sequences to changes in digital artefacts and communication exchanges, indicating that iterations were associated with incomplete information or differences in interpretation across roles. The findings show that combining process mining with contextual inquiry enables the identification of iterations and clarifies the conditions under which they emerge.
- Open Access
- 27 January 2026
- Published by MDPI AG in Preprints.org
Process Mining in Digital Dental Laboratories: Identifying Iterations Through Actions and Digital Artefacts
Digitalization has reshaped dental laboratory processes, introducing new tools and artefacts that support collaboration between clinicians, technicians, and engineers. However, these processes still involve collaboration challenges and frequent iterations that extend delivery times and complicate communication. This study aimed to analyze how digital processes unfold in a dental laboratory and to identify the causes of iterations during the design and manufacturing of custom abutments. Ten cases were selected for analysis, and their event logs were analyzed using process mining in Disco. The analysis reconstructed the process from case initiation to delivery, revealing a reference process consisting of scanning, CAD design, verification, manufacturing, and quality control. Iterations most frequently occurred during CAD and verification phase, often triggered by incomplete clinical information, differing interpretations, or design inconsistencies. Artefacts such as order forms, 3D models, and design outputs were found to mediate collaboration but also to introduce new dependencies and sources of rework. The findings show that variations in digital processes are closely tied to communication and artefact interpretation rather than technical errors. The study demonstrates how process mining can support exploratory analysis of observed process cases derived from dental event logs and provides a basis for earlier recognition of information gaps that contribute to repeated or newly introduced phases within design and manufacturing processes in digital dentistry.
- 27 January 2026
- Published by MDPI AG in Preprints.org
Open AccessProcess Mining in Digital Dental Laboratories: Identifying Iterations Through Actions and Digital Artefacts
Digitalization has reshaped dental laboratory processes, introducing new tools and artefacts that support collaboration between clinicians, technicians, and engineers. However, these processes still involve collaboration challenges and frequent iterations that extend delivery times and complicate communication. This study aimed to analyze how digital processes unfold in a dental laboratory and to identify the causes of iterations during the design and manufacturing of custom abutments. Ten cases were selected for analysis, and their event logs were analyzed using process mining in Disco. The analysis reconstructed the process from case initiation to delivery, revealing a reference process consisting of scanning, CAD design, verification, manufacturing, and quality control. Iterations most frequently occurred during CAD and verification phase, often triggered by incomplete clinical information, differing interpretations, or design inconsistencies. Artefacts such as order forms, 3D models, and design outputs were found to mediate collaboration but also to introduce new dependencies and sources of rework. The findings show that variations in digital processes are closely tied to communication and artefact interpretation rather than technical errors. The study demonstrates how process mining can support exploratory analysis of observed process cases derived from dental event logs and provides a basis for earlier recognition of information gaps that contribute to repeated or newly introduced phases within design and manufacturing processes in digital dentistry.
- Open Access
- 27 August 2025
- Published by Cambridge University Press (CUP) in Proceedings of the Design Society
- Vol. 5, 1585-1594
Evaluating ChatGPT’s role in collaborative CAD task completeness
This study explores the role of ChatGPT in the completeness of collaborative computer-aided design (CAD) tasks requiring varying types of engineering knowledge. In the experiment involving 22 pairs of mechanical engineering students, three different collaborative CAD tasks were undertaken with and without ChatGPT support. The findings indicate that ChatGPT support hinders completeness in collaborative CAD-specific tasks reliant on CAD knowledge but demonstrates limited potential in assisting open-ended tasks requiring domain-specific engineering expertise. While ChatGPT mitigates task-specific challenges by providing general engineering knowledge, it fails to improve overall task completeness. The results underscore the complementary role of AI and human knowledge.
- 27 August 2025
- Published by Cambridge University Press (CUP) in Proceedings of the Design Society
- Vol. 5, 1585-1594
Open AccessEvaluating ChatGPT’s role in collaborative CAD task completeness
This study explores the role of ChatGPT in the completeness of collaborative computer-aided design (CAD) tasks requiring varying types of engineering knowledge. In the experiment involving 22 pairs of mechanical engineering students, three different collaborative CAD tasks were undertaken with and without ChatGPT support. The findings indicate that ChatGPT support hinders completeness in collaborative CAD-specific tasks reliant on CAD knowledge but demonstrates limited potential in assisting open-ended tasks requiring domain-specific engineering expertise. While ChatGPT mitigates task-specific challenges by providing general engineering knowledge, it fails to improve overall task completeness. The results underscore the complementary role of AI and human knowledge.
- Open Access
- 27 August 2025
- Published by Cambridge University Press (CUP) in Proceedings of the Design Society
- Vol. 5, 3251-3260
Uncovering collaboration dynamics in design projects using network analysis and log data
This study explores the integration of network analysis and CAD/PDM log data to analyze collaboration and activity patterns in a multi-year engineering project. Using logs from a collaborative CAD platform with PDM features, the research examines team interactions and network evolution over time. Key findings reveal that early project stages featured smaller, denser networks, while later stages saw larger, less interconnected structures. Subteam formations were dynamic, with variations in size and number. Individual-level analysis showed that user influence, measured through eigenvector centrality, did not always align with activity volume. This work highlights the potential of CAD/PDM data for understanding collaboration dynamics and lays the groundwork for further studies on team interactions in design processes.
- 27 August 2025
- Published by Cambridge University Press (CUP) in Proceedings of the Design Society
- Vol. 5, 3251-3260
Open AccessUncovering collaboration dynamics in design projects using network analysis and log data
This study explores the integration of network analysis and CAD/PDM log data to analyze collaboration and activity patterns in a multi-year engineering project. Using logs from a collaborative CAD platform with PDM features, the research examines team interactions and network evolution over time. Key findings reveal that early project stages featured smaller, denser networks, while later stages saw larger, less interconnected structures. Subteam formations were dynamic, with variations in size and number. Individual-level analysis showed that user influence, measured through eigenvector centrality, did not always align with activity volume. This work highlights the potential of CAD/PDM data for understanding collaboration dynamics and lays the groundwork for further studies on team interactions in design processes.
- 8 April 2025
- Published by Elsevier in Expert Systems with Applications
- Vol. 282
Efficient, effective, or both? The effect of immersion and goal on design review team performance
- 8 April 2025
- Published by Elsevier in Expert Systems with Applications
- Vol. 282
Efficient, effective, or both? The effect of immersion and goal on design review team performance
- Open Access
- 8 November 2024
- Published by MDPI AG in Education Sciences
- Vol. 14 (11) , 1226
Towards Computer-Supported Functional Modelling in Engineering Design Education
The growing need for solutions that can support the computer-based and distant assessment of functional models has resulted in ad hoc implementations of various diagramming tools. These tools are typically not intended for the purpose of functional modelling and lack the flexibility and efficiency of the traditional pen-and-paper approach. This paper reports on an experimental study of 42 students who were introduced to functional modelling through either printed vocabulary materials for pen-and-paper modelling or a specifically developed software application for computer-based modelling. All participants received an identical task—model an electric citrus juicer—with a brief description of how one operates and a photograph of a commercial example. The results show no significant difference in their total scores. However, the pen-and-paper group performed significantly better when it came to the selection of appropriate functions and creating plausible function–flow pairs. These results suggest that the current version of the software alters the functional modelling process in which the students typically engage. Also, it has been hypothesised that the software tool’s lack of flexibility and dynamism in presenting the predefined function vocabulary, when compared to the traditional printouts, might result in earlier fixation and the selection of less appropriate functions. On the other hand, the computer-supported approach can be better controlled and is less prone to critical errors, such as disregarding functional modelling conventions.
- 8 November 2024
- Published by MDPI AG in Education Sciences
- Vol. 14 (11) , 1226
Open AccessTowards Computer-Supported Functional Modelling in Engineering Design Education
The growing need for solutions that can support the computer-based and distant assessment of functional models has resulted in ad hoc implementations of various diagramming tools. These tools are typically not intended for the purpose of functional modelling and lack the flexibility and efficiency of the traditional pen-and-paper approach. This paper reports on an experimental study of 42 students who were introduced to functional modelling through either printed vocabulary materials for pen-and-paper modelling or a specifically developed software application for computer-based modelling. All participants received an identical task—model an electric citrus juicer—with a brief description of how one operates and a photograph of a commercial example. The results show no significant difference in their total scores. However, the pen-and-paper group performed significantly better when it came to the selection of appropriate functions and creating plausible function–flow pairs. These results suggest that the current version of the software alters the functional modelling process in which the students typically engage. Also, it has been hypothesised that the software tool’s lack of flexibility and dynamism in presenting the predefined function vocabulary, when compared to the traditional printouts, might result in earlier fixation and the selection of less appropriate functions. On the other hand, the computer-supported approach can be better controlled and is less prone to critical errors, such as disregarding functional modelling conventions.
- Open Access
- 16 May 2024
- Published by Cambridge University Press (CUP) in Proceedings of the Design Society
- Vol. 4, 45-54
Virtual design hackathons: a data collection framework
Design hackathons offer a unique research opportunity to study time-pressured collaborative design. At the same time, research on design hackathons faces unique methodological challenges, prompting the exploration of new research approaches. This paper proposes a new data-collection framework that leverages a virtual format of hackathon events and enables a deeper insight into hackathon dynamics. The framework applicability is presented through a case study of the IDEA challenge hackathon, in which different intrusive and non-intrusive data collection approaches were used.
- 16 May 2024
- Published by Cambridge University Press (CUP) in Proceedings of the Design Society
- Vol. 4, 45-54
Open AccessVirtual design hackathons: a data collection framework
Design hackathons offer a unique research opportunity to study time-pressured collaborative design. At the same time, research on design hackathons faces unique methodological challenges, prompting the exploration of new research approaches. This paper proposes a new data-collection framework that leverages a virtual format of hackathon events and enables a deeper insight into hackathon dynamics. The framework applicability is presented through a case study of the IDEA challenge hackathon, in which different intrusive and non-intrusive data collection approaches were used.
- Open Access
- 19 June 2023
- Published by Cambridge University Press (CUP) in Proceedings of the Design Society
- Vol. 3, 1377-1386
DESIGNING FOR LEGAL PRACTITIONERS: LESSONS LEARNED FROM LEGAL TECH DEVELOPMENT AND IMPLEMENTATION
Recently, the design and digitalisation approaches have become increasingly utilised in the legal context, typically under the names of legal design and legal tech. One of their goals is to help legal practitioners be more efficient and to provide better quality and more comprehensive legal services. Also, given that both movements rely heavily on participatory and co-design, they will require increased support not only from design practitioners but also from design researchers and educators. Therefore, this paper investigates, from a design research viewpoint, the opportunities and challenges of developing and implementing legal tech, with a particular focus on legal practitioners. It reports on four cases of designing legal tech solutions and their implementation in a law firm. The main insights are related to the importance of value perception through participatory and co-design, the need for efficient and effective testing methodologies, and the opportunity to test a wide range of design methods and tools in the legal context. The paper also complements the legal design and legal tech literature with additional arguments on why designing in the legal context is challenging compared to designing in other domains.
- 19 June 2023
- Published by Cambridge University Press (CUP) in Proceedings of the Design Society
- Vol. 3, 1377-1386
Open AccessDESIGNING FOR LEGAL PRACTITIONERS: LESSONS LEARNED FROM LEGAL TECH DEVELOPMENT AND IMPLEMENTATION
Recently, the design and digitalisation approaches have become increasingly utilised in the legal context, typically under the names of legal design and legal tech. One of their goals is to help legal practitioners be more efficient and to provide better quality and more comprehensive legal services. Also, given that both movements rely heavily on participatory and co-design, they will require increased support not only from design practitioners but also from design researchers and educators. Therefore, this paper investigates, from a design research viewpoint, the opportunities and challenges of developing and implementing legal tech, with a particular focus on legal practitioners. It reports on four cases of designing legal tech solutions and their implementation in a law firm. The main insights are related to the importance of value perception through participatory and co-design, the need for efficient and effective testing methodologies, and the opportunity to test a wide range of design methods and tools in the legal context. The paper also complements the legal design and legal tech literature with additional arguments on why designing in the legal context is challenging compared to designing in other domains.
- Open Access
- 28 July 2021
- Published by U-turn Press LLC in Computer-Aided Design and Applications
- Vol. 19 (2) , 320-335
Identifying Indicators for the Use of Virtual Prototypes in Distributed Design Activities
- 28 July 2021
- Published by U-turn Press LLC in Computer-Aided Design and Applications
- Vol. 19 (2) , 320-335
Open AccessIdentifying Indicators for the Use of Virtual Prototypes in Distributed Design Activities