This special issue of Macromolecular Reaction Engineering is dedicated to “Statistical Modeling Tools and Approaches for Polymerization Reaction Engineering” and is a continuation of the special issue published in 2014,1 dedicated to “Mathematical Tools and Approaches for Polymerization Reaction Engineering.” After finishing that first volume, we quickly realized that some important aspects of statistical tools frequently employed in the polymerization field had not been discussed in sufficient detail in the interesting papers submitted by our colleagues at that opportunity. As a consequence, even before publishing the first volume of this series, we decided to start the preparation of this second volume of this series. As in the previous case, I was very glad when Dr. Spiegel decided to support the preparation of this special issue of Macromolecular Reaction Engineering and invited me again to help him with the organization of this volume. As a matter of fact, I was captured and became an enthusiastic supporter of statistical methods and approaches in the 1980s, when I was a graduate student at COPPE/UFRJ and found out that statistical experimental design tools2-4 could allow for significant reduction of experimental costs and maximization of the information content of experimental data obtained with very hard work at research labs. Since then, I have been flirting with statistical methods for more than 30 years so that I could not avoid the temptation to get involved in this very interesting project. Statistical tools and approaches have been used in the polymerization field for many decades, as brilliant researchers, such as W. H. Carothers and J. P. Flory, realized that polymerization reaction mechanisms could not be entirely described by deterministic rules, as polymer materials can almost always be regarded as random mixtures of macromolecules that present distinct sizes, compositions, degrees of branching, and so on.5-7 Although reproducible, the final properties of most polymer materials prepared inside the usually very different reaction vessels and processes result in all cases from a complex network of random fundamental reaction steps that take place simultaneously, as a magnificent orchestra that emerges from the apparent chaos. In this scenario, it seems reasonable to assert that one of the most important roles of scientists and engineers who work in this particular field is the understanding and the manipulation (or designing) of the individual probabilities that control the occurrence of the respective reaction events in such complex mechanisms. Although statistical events are intrinsically connected to the hearts and souls of polymerization reaction mechanisms, the use of statistical tools and approaches in the polymer field can also be related to other applications. For example, the existence of complex reaction networks usually leads to large number of model parameters, which must be inferred from available data. Due to unavoidable existence of experimental uncertainties, model parameters must then be analyzed carefully with help of consistent statistical tools.8 Besides, as statistical analyses usually rely on large number of independent simulations, implementation of statistical procedures requires proper development and implementation of advanced computational schemes in order to provide useful results in the proper time.9 For this reason, the fast development of computer resources has also encouraged the continuous development of statistical applications in the polymer field. Based on the previous paragraphs, the papers published in the current volume consider applications of statistical modeling and tools in polymerization processes for different purposes and pursuing different objectives. For instance, the paper by Cui et al.10 discusses the estimation of model parameters from experimental data obtained during the production of poly(trimethylene ether glycol) from 1,3-propanediol. Statistical methods are required for formulation of the parameter estimation problem, determination of parameter significance, and interpretation of model adequacy, as usual in the polymer field. The papers by Scott et al.11 and Kazemi et al.12 are devoted to statistical design of experiments for optimal estimation of kinetic parameters in polymerization models. In the first case, D-optimum designs are proposed to allow for maximization of the information content of model parameters and for reduction of the model size. As shown by the authors, the use of statistical designs can lead to more precise model parameters and more compact model formulations. In the second case, the authors use the Fisher information matrix to design experiments for improved estimation of reactivity ratios in copolymerization problems, taking into account the intrinsic variability of measured variables. As shown by the authors, the proposed approach indicates that the experimental ranges that lead to optimum estimation of reactivity ratios are located in the vicinities of the corners of the terpolymerization composition triangular plot. The paper by Tobita13 makes use of Markovian chain approaches and Monte Carlo simulations in order to analyze the effects of chain branching and scission kinetics on the branched structure of low density polyethylenes. Particularly, it is shown that the effect of different kinetics on the polymer structure can be negligible when the final monomer conversion is smaller than 25%, normally used for the commercial production processes, but can be very significant if conversion becomes higher. Monte Carlo methods were also employed by Lemos et al.14 to analyze the effect of residence time distributions on the molecular weight distributions and chain composition distributions of copolymers produced in tubular reactors through controlled radical polymerization mechanisms. The proposed technique assumes that the analyzed system can be described as a set of batch reactors operated independently, whose volumes and batch times can be related to the discretized version of the residence time distribution. Pladis et al.15 proposed the use of a modified kinetic Monte Carlo method for analyses of the viscoelastic properties of low-density polyethylenes produced in high pressure tubular reactors. The proposed model uses an ensemble of branched polymer chains generated by the Monte Carlo procedure to describe the viscoelastic properties of the obtained product, which are compared to real experimental measurements of grades produced in industrial-scale reactors. As one can observe, Monte Carlo methods are so important for statistical modeling of polymerization processes and polymer properties that the field is reviewed thoroughly by Brandão et al.16 As discussed by the authors, the use of Monte Carlo methods will probably become even more important in the future, given the fast development of computer resources and the high flexibility provided by the method to describe different processes and materials. In spite of that, Kryven and Iedema17 show that deterministic procedures based on population balances can also provide competitive model structures, both in terms of computer costs and model responses and when compared to Monte Carlo methods, if appropriate description of the reaction mechanism and kinetics is provided and suitable numerical schemes are implemented. We sincerely hope the readers of Macromolecular Reaction Engineering will appreciate this second volume of papers concerned with modeling and numerical aspects of mathematical methods used in the polymerization field. We also hope this collection of papers will stimulate the increasing use of statistical methods in this field. Meanwhile, we will keep sending Stefan a steady flow of random perturbations in order to create the appropriate momentum for a third volume of this series, most likely on computational fluid dynamics aspects. José Carlos Pinto obtained his BSc in Chemical Engineering at Universidade Federal da Bahia (Salvador, Bahia, Brazil) in 1985, his MSc in Chemical Engineering at Universidade Federal do Rio de Janeiro (Rio de Janeiro, Rio de Janeiro, Brazil) in 1987, and his DSc in Chemical Engineering at Universidade Federal do Rio de Janeiro in 1991. At present, he holds a position of Professor Titular at Programa de Engenharia Química/COPPE, Universidade Federal do Rio de Janeiro, is Professor Permanente at Programa de Pós-Graduação em Química, Instituto Militar de Engenharia, and is a full-member of the Academia Brasileira de Ciências and Academia Nacional de Engenharia. José Carlos has worked in the general field of modeling, simulation, and control of polymerization processes since 1987, has published about 300 papers in refereed journals, and has deposited 30 patents. José Carlos has been the coordinator of more than 100 projects with industrial partners and has advised more than 100 MSc Dissertations and 50 DSc Theses.