Digital twins are increasingly used in engineering to replicate and predict machine and process behavior. However, existing digital twin implementations for software‑in‑the‑loop simulation of industrial food‑processing machinery remain limited in their ability to model the mechanical response of food materials during processing. This limitation becomes particularly apparent for natural, heterogeneous products such as fish, where most processing steps involve cutting soft biological tissue. Consequently, there is a need for computationally efficient modeling approaches that can accurately represent soft tissue mechanics, including damage and fracture. In this work, we introduce a general particle‑based simulation framework explicitly designed to incorporate material damage and fracture, similar to [1]. Within this framework, we examine two distinct approaches for representing material behavior: (i) a formulation based on nonlinear springs connecting discrete particles, and (ii) an approach using deformation‑gradient estimation to capture continuum‑like responses. Both methods are evaluated with respect to their ability to reproduce established material laws. To demonstrate their validity, we compare simulations against analytical solutions for uniaxial deformation as well as benchmark results obtained from finite element analysis. Furthermore, we relate the two formulations to peridynamics theory and their specific stabilization technique [2] and demonstrate, how a generalized peridynamics‑based framework can be used for the same purpose. Building on this theoretical foundation, we derive a simple parameterizable material model for soft fish tissue, which is based on the results of uniaxial compression tests conducted on post‑rigor Atlantic salmon (Salmo salar) tissue. In conclusion, we discuss the potential of the proposed framework and corresponding simulation strategies as a basis for digital twins in applications involving deformable biological materials and cutting operations. The presented methods provide a promising step toward more realistic and predictive virtual models for industrial food‑processing systems. This research work was supported by the SGS-2025-015 project of the University of West Bohemia.