To better study biological effects of space radiation using ground-based facilities, the NASA Space Radiation Laboratory (NSRL) at the Brookhaven National Laboratory has been upgraded to rapidly switch ions and energies. This has allowed investigators to design irradiation protocols comprising a mixture of ions and energies more indicative of the galactic cosmic ray (GCR) environment. Despite these advancements, beam selection and delivery schemes should be optimized against facility and experimental constraints and validated to ensure such irradiations are a suitable representation of the space environment. Importantly, since experiments are time consuming and expensive, models capable of predicting biological outcomes over a range of irradiation conditions (single ion, sequential multi ion or mixed fields) are needed to support such efforts. In this work, human fibroblasts were placed behind 20 g/cm2 aluminum and 10.345 g/cm2 polyethylene and irradiated separately by 344 MeV hydrogen, 344 MeV/n helium, 450 MeV/n oxygen and 950 MeV/n iron ions at various doses. The fluorescence in situ hybridization (FISH) whole chromosome painting technique was then used to assess the cells for chromosome aberrations (CAs), notably simple exchanges. A multi-scale modeling approach was also developed to predict the formation of chromosome aberrations in these experiments. The Geant4 simulation toolkit was used to determine the spectra of particles and energies produced by interactions between the incident beams and shielding. The simulated mixed field generated by shielding was then transferred into the track structure code, RITRACKS (relativistic ion tracks), to generate three-dimensional (3D) voxelized dose maps at the nanometer scale. Finally, these voxel dose maps were input into the new damage and repair model, RITCARD (radiation-induced tracks, chromosome aberrations, repair and damage), to predict the formation of various CAs. The multi-scale model described herein is a significant advancement for the computational tools used to predict biological outcomes in cells exposed to highly complex, mixed ion fields related to the GCR environment. Results show that the simulation and experimental data are in good agreement for the complex radiation fields generated by all ions incident on shielding for most data points. The differences between model predictions and measurements are discussed. Although improvements are needed, the model extends current capabilities for evaluating beam selection and delivery schemes at the NSRL ground-based GCR simulator and for informing NASA risk projection models in the future.