0603316 20250603151638.420250102235959.9 85212574810 001392471600001 10.1016/j.compstruc.2024.107630 12 s. E cav_un_epca*02564490045-7949307Elsevier Modular-topology optimization Reusability Free-material optimization Additive manufacturing Modularity Compliant mechanism cav_un_auth*0454762 Tyburec Marek UTIA-B Matematická teorie rozhodování Department of Decision Making Theory MTR MTR CZ 35 A Ústav teorie informace a automatizace AV ČR, v. v. i. cav_un_auth*0479359 Doškář M. CZ 25 cav_un_auth*0371413 Somr M. CZ 20 cav_un_auth*0409618 Kružík M. CZ cav_un_auth*0018366 Zeman J. CZ 10 https://library.utia.cas.cz/separaty/2025/MTR/tyburec-0603316.pdf https://www.sciencedirect.com/science/article/pii/S0045794924003596?via%3Dihub GX19-26143X GA ČR CZ cav_un_auth*0440774 L100752301 GA AV ČR CZ cav_un_auth*0479360 Modular designs have gained popularity because they can generally address manufacturing efficiency, reusability, and sustainability concerns. Here, we contribute to the growing field by proposing a fully automatic design method for modules utilized in several products. Our manufacturing-aware procedure is composed of three consecutive steps: (i) free-material optimization for obtaining the optimal spatially distributed elasticity tensors, (ii) hierarchical clustering of the stiffness tensors directly into a given number of modules while allowing for their flipping, and (iii) single-scale topology optimization with manufacturing aspects to design the final topology of mechanically compatible modules. These aspects include connectivity constraints to ensure the integrity of all modules and the three-field projection to account for manufacturing inaccuracies. We illustrate the entire procedure with the design and fabrication of eight different reusable modules assembled into well-functioning modular inverter and gripper mechanisms. These mechanisms were 3D printed and subjected to mechanical testing using an in-house testing machine and digital image correlation. The experimental results show excellent agreement between the predicted and observed behavior and highlight the potential of the method for scalable additive manufacturing. 2026 JI WOS 20000 20100 20101 5 RVO:67985556 https://hdl.handle.net/11104/0363654 cav_un_auth*0300409 České vysoké učení technické v Praze, Stavební fakulta ČVUT, FSv CZ S 107630 C ENGINEERING.CIVIL|COMPUTERSCIENCE.INTERDISCIPLINARYAPPLICATIONS 4.9 1.2 13.6 0.00602 1.035 16917 168 1.296 57.52 5.32 2419 Q1 4.400 235 Q1 78.6 1 Q1 83.4 2025 85212574810 SCOPUS PUBMED 001392471600001 WOS cav_un_epca*0256449 Computers and Structures 0045-7949 1879-2243 Roč. 307 č. 1 2025 Elsevier