Optimal experimental design for the calibration of a high-temperature thermal strain model for concrete during coolingPut, Florian (Korespondenčni avtor) Sørensen, Matilde Bruun (Avtor) Abbiati, Giuseppe (Avtor) Lucherini, Andrea (Avtor) Merci, Bart (Avtor) Van Coile, Ruben (Avtor) optimal experimental designconcretethermal straincoolingPerformance-based structural fire design relies on models that capture material and structural behaviour during heating and cooling. Such models require experimental data, but experiments are often time- and resource- intensive. Optimal Experimental Design (OED) can reduce the number of tests needed by minimizing the variance of parameter estimates. This study demonstrates the use of OED, using D-optimality as the optimization criterion, for an experimental setup that measures the thermal elongation of concrete specimens. In these tests, cylindrical concrete specimens are slowly heated to a predefined maximum temperature while their elongation is being measured. The goal of the experimental campaign is to calibrate a model for the free thermal strain of concrete during cooling. The OED determines the optimal exposure that is expected to result in the lowest variance around the mean values of the parameter estimates. The results of the OED are compared with a baseline experimental design without optimization, showing that the advantages of OED become increasingly evident as the number of experimental runs grows and intuitive reasoning becomes less reliable. In addition, the approach is validated considering real experimental data.Elsevier20262026-04-08 10:52:33Neznano28808sl© 2026 Elsevier Ltd.