The risk of thermal runaways is a major concern in many industrial polymerizations. Thermal runaways are the leading cause of plant explosions globally, resulting not only in significant chemical spills but also in civilian casualties. The combination of high values of polymerization heats of reaction (100-200 kJ/mol) and apparent activation energies (well above 40 kJ/mol) leads to positive thermal feedback in polymerization reactors. In other words, a minor increase in temperature results in a significant rise in the reaction rate and the heat generated. This FWO proposal focuses on temperature regulation at any reactor scale to minimize the risk of runaway reactions in polymerization processes. The ultimate objective of the project is to enhance our understanding of how process variables, spatial heterogeneity, and reactor configurations affect the process safety of polymer production and the final polymer properties. To achieve this, we will link a more realistic and fundamental representation of the macroscale phenomena occurring in industrial-scale reactors with the detailed molecular output that the state-of-the-art Coupled Matrix-based Monte Carlo technology, as pioneered by LCT, is already capable of predicting. To describe the macroscale heterogeneity, we will divide the chemical reactor into compartments, each representing several distinct reaction regions. This approach will be validated using available data from industrially relevant case studies.