To improve the durability and performance of proton exchange membrane fuel cells, we present a quasi-2D, time-dependent multiphase model that captures key spatial and temporal dynamics with high computational efficiency. Unlike 0D/1D models, it resolves spatial variations, and unlike full 3D models, it enables fast transient simulations. The model incorporates convection, diffusion, capillary effects, membrane water transport, nitrogen crossover, heat generation, and sorption kinetics. A linearisation scheme with Chebyshev collocation ensures near real-time performance. Validation against detailed 3D CFD shows strong agreement in polarisation behaviour, species distributions, water content, and temperature profiles. Simulations under dynamic loads highlight the importance of spatially resolved water transport. This efficient, accurate model enables advanced control design, parametric studies, and predictive diagnostics, supporting improved fuel cell operation and lifetime.