https://unilib.phaidrabg.rs/o:8605 doi:10.46793/41DAS2025.071R ISBN: 978-86-6335-157-8 soil constitutive models, automated parameter identification procedure, Modified Cam-Clay model (MCC) http://creativecommons.org/licenses/by-nc/4.0/legalcode info:eu-repo/semantics/conferenceProceedings Rakić, Dragan Radovanović, Slobodan Živković, Miroslav application/pdf 27129026 bytes 41st Danubia-Adria Symposium Advances in Experimental Mechanics - Proceedings, September 23-26, 2025, Kragujevac, Serbia startpage: 71 endpage: 74 For reliable stability analysis of geotechnical structures, accurate calibration of constitutive model parameters is essential. These parameters govern the simulated material response under load, and any uncertainty can lead to significant deviations in predicted behavior. Conventional parameter determination relies on experimental testing of soil samples, whereby stress–strain measurements are used to infer individual model parameters according to their constitutive definitions. However, this manual fitting process is complex and time-consuming, particularly for soils, which exhibit highly nonlinear behavior. To address these challenges, we propose an automated parameter identification procedure for constitutive models. The methodology is developed for the Modified Cam-Clay model (MCC), a widely adopted framework for simulating the mechanical response of soft clays and normally consolidated soils. The first part of this paper presents the theoretical foundations and the implicit stress integration algorithm for the MCC model, implemented within the PAK finite element software. For automated identification, the same integration scheme is translated into Python and applied at the level of single integration point. This point-wise approach is justified for homogeneous stress states, such as those encountered in standard laboratory tests (e.g., oedometer and triaxial tests). In the second part of the paper, we describe the parameter identification program, which interfaces with the Python integration routine to perform optimization against experimental data. Finally, the developed identification algorithm is verified through a comparison of parameter estimates obtained from the PAK-based finite element implementation and those produced by the automated procedure. Faculty of Engineering University of Kragujevac, Kragujevac 2025 Automation of Constitutive Model Parameter Identification eng