Low-density polyethylene (LDPE) is a versatile polymer which is widely used both in domestic and industrial applications. About 50% of commercial LDPE production around the world is carried out by free-radical polymerization initiated by organic peroxides in high-pressure tubular reactors. The free-radical polymerization is highly sensitive to even small changes in process conditions. As a result, the extreme pressures and temperatures in the high-pressure tubular LDPE reactor (HPTLR) present operational challenges to the plant engineer. A reliable model of the HPTLR will prove to be a valuable tool for the plant engineer in making decisions about reactor operations;In this work, a comprehensive dynamic model for the HPTLR has been developed for the prediction of reactor behavior. The model combines the power of CFD (computational fluid dynamics)-based modeling with the simplicity and intuition of CRE (chemical reaction engineering)-based micromixing models. The mathematical model is based on the finite-mode PDF (probability density function) approach to modeling turbulent reactive flows. The advantage of this approach is that the source terms in the model transport equations appear in closed form thus obviating the need for closure approximations which are common in (and the main shortcoming of) CRE-based reactor models. Some simplifying assumptions (which are found to be true in most cases of HPTLR operations) have been made, important among which is the assumption of fully-developed turbulent pipe flow. A two-environment model is used to describe micromixing. Free-radical mechanisms are considered for both LDPE polymerization reactions as well as ethylene decomposition reactions. Heat-transfer effects between the reactor tube and the coolant water jacket surrounding it are also considered. The resultant transport equations, which are convection-diffusion-reaction (CDR) - type partial differential equations, are numerically solved using a fractional time-stepping method;Model parameters have been estimated using plant data for both steady-state and dynamic behavior. Using experimental design techniques, the most influential parameters in the model have been determined. Dynamic behavior of the model under typical reactor operating conditions was studied. From our studies, we conclude that our model should prove to be an effective tool for the plant engineer in making critical plant-operation decisions.