A predictive model to determine the rollover threat of heavy vehicles is proposed. The purpose of this model is to predict the rollover threat sufficiently in advance of the actual event so as to enable the driver to react accordingly. The Load Transfer Ratio (LTR) is used as a rollover threat index in this model. In order to facilitate a practical implementation of the proposed approach, a predictive model is established using a simple roll-plane model of the vehicle in conjunction with an online parameter identification strategy. This is to address the inherent issues involved in assessing certain key parameters in the proposed predictive model. The proposed predictive model and the associated parameter-identification algorithm are verified using a 14 degrees-of-freedom vehicle simulation model previously developed as a part of a broader study on heavy vehicles undertaken at Texas A&M University and supported by Texas Transportation Institute (TTI).