This paper discusses an integrated thermal-hydraulic (TH) uncertainty analysis methodology with an application to the Loss-of-Fluid Test (LOFT) test facility large-break loss-of-coolant accident (LBLOCA) transient. The methodology is intended for applications to best-estimate analyses of complex TH codes. The goal is to develop an integrated method to make such codes capable of comprehensively supporting the uncertainty assessment with the ability to handle important accident transients. The proposed methodology considers the TH code structural uncertainties (generally known as model uncertainty) explicitly by treating internal submodel uncertainties and by propagating such model uncertainties in the code calculations, including uncertainties about input parameters. The methodology is probabilistic, using the Bayesian approach for incorporating available evidence in quantifying uncertainties in the TH code predictions. The types of information considered include experimental data, expert opinion, and limited field data, in treating both model and input parameter uncertainties. The code output is further updated through additional Bayesian updating with available experimental data from the integrated test facilities. The methodology uses an efficient Monte Carlo sampling technique for the propagation of uncertainty, in which a modified Wilks' sampling criteria of tolerance limits is used to significantly reduce the number of simulations. This paper describes the key elements of the uncertainty analysis methodology and summarizes its application to the LOFT test facility LBLOCA.