The determination of the BMD involves conducting dose-response experiments where animals or cell cultures are exposed to varying concentrations of a substance. The data collected from these experiments are then analyzed using statistical models to fit a dose-response curve. Several types of models can be used, such as probit, logistic, and Weibull models. The BMD is the dose at which a benchmark response level, usually a small increase in the incidence of an adverse effect, is observed. The Benchmark Dose Lower Confidence Limit (BMDL) is often used as a conservative measure for regulatory purposes.