Modern automotive battery systems require coupled electro-thermal-mechanical simulations of lithium-ion cells and packs to quantify cycling-induced residual stresses as well as swelling and deformation behavior, which are key contributors to the state of health. The thermal conditions and resulting temperature fields in operational scenarios are driven by electrically generated losses and therefore depend on charge/discharge rate and the applied cooling boundary conditions. While electro-thermal modeling, fluid-based flow analysis, and structural simulations are well established individually, their integration into a unified computational framework applicable to industrial-scale battery packs with complex geometries remains subject of current research. This contribution presents a multi-physics coupling approach in which electrical loss modeling, thermal conduction, cooling by incompressible CFD, and nonlinear structural mechanics are integrated and applied to an industry representative prismatic cell and subsequently to a full battery pack. Electrical behavior is represented using a Randles type equivalent circuit model, with resistive losses mapped as heat sources to the thermal domain. Heat transport in solid domains (cells, housing, interfaces) is solved concurrently with the heat transport in the fluid domain, where the fluid-solver resolves the conjugate heat transfer. In particular, the fluid approach first solves a stand-alone steady-state flow problem. The resulting flow field is then used to compute the heat-transfer solution for the complete assembly. The subsequent thermal analysis is performed on a different fluid mesh using mapping capabilities, without rerunning the steady-state flow simulation. The coupled treatment enables comparative analysis of different configurations and boundary conditions, including internal stresses during battery cycle loads and temperature propagation effects. A parameter study shows that the predicted stress evolution varies significantly depending on whether cooling is considered and on the specific thermal conditions applied. The study demonstrates the relevance of multi-physics-based approaches for accurately representing operational scenarios in practical applications.