Many high performance products are made with thermoplastic/fiber composites by injection molding processes. The shear forces induced in injection molding produce fibers breakage which detriments the final mechanical properties of the material. Therefore, the prediction of the evolution of the topology of the fibers during the molding process is a critical aspect for obtaining an optimal final product. The runners of the injection molds are the critical zones where fiber breakage takes place. The fiber breakage process inside the mold runners can be more easily studied from capillary extrusion. The fibers can be characterized as a population of entities that break as they move along the capillary. As such, the population balance equation (PBE) can be used to study the breakage of the fibers. This equation makes a statistical description of the evolution of a group of entities. The PBE uses a density function as its main variable, defined as function of time, physical coordinates, and internal coordinates. The latter are used to represent properties of the entities, such as size, mass, temperature, composition, etc.. In the case of fiber capillary extrusion, the density function is affected by breakage events. The aim of this work is to analyze the applicability of the PBE for predicting the evolution of the fiber breakage in polymer/short-fiber composite during capillary extrusion. The least squares spectral method will be used to solve the resulting integro-differential PBE equation. The main goal of this work is to show the evolution of a density function predicted with different breakage kernels and to validate the modeling results with experimental data. Four intrinsically different breakage kernels and two redistribution functions are tested.