Stochastic characterization of municipal solid waste properties for thermochemical applications

Heterogeneity in municipal solid waste (MSW) properties remains a major challenge for thermochemical processing applications in both energy recovery and recycling sectors. The inherently heterogeneous nature of waste directly affects its characteristics as feedstock, complicating process design, operational control, and numerical modeling. The present study develops a probabilistic framework for the hierarchical classification and aggregation of waste properties. Waste is assumed to be composed of four main groups: Paper, Organic, Plastic, and Inert. Each of them is divided into representative subgroups. Thermophysical properties, including specific density, heat capacity, and thermal conductivity, together with compositional data as proximate and elemental analyses, are evaluated to quantify the influence of the different main groups on the overall waste mixture behavior. The framework is based on an extensive database gathered from literature considering the four main groups and 33 subgroups, including 1259 data points for thermophysical properties, 346 for proximate analysis, 469 for elemental analysis of combustible fractions, and 120 for chemical characterization of inert components. The method conserves the intrinsic variability of the dataset, restricts the introduction of additional parametric assumptions during aggregation, and ensures a consistent and traceable transfer of uncertainty across the different levels. Main group influence on the generated waste samples is analyzed and discussed. The resulting probabilistic description supports more realistic numerical simulations, facilitates sensitivity analyses, and enhances robustness in thermochemical process design, operation and optimization.