Assessment of the bioequivalence (BE) between the test drug and patent drug is a major issue to the biopharmaceutical industry for marketing the test drug as a generic drug. To do so, a 2х2 crossover design is usually conducted for a pharmacokinetic (PK) study where the bioavailability (BA) parameters such as area under the concentration-time curve (AUC) or maximum concentration (Cmax), are employed in a BE test. In this paper, we consider mixed-effect models for the estimated BA parameters where Student's t-distribution is used to describe the between-subject variation and measurement error variables. Based on the proposed robust model, we suggest a Bayesian test for the BE of the two drugs. A Monte Carlo study is further conducted to investigate the posterior probability of bioequivalence associated with different models under different prior distributions of the BA parameters. A real data set is finally illustrated by using the proposed model and test.