A Coupled D3Q19-LBM and Social Force Framework for Indoor Fire Evacuation Under Smoke, Heat and Ventilation Effects

To characterise the coupled effects of smoke dispersion, temperature rise, ventilation-driven flow and pedestrian evacuation behaviour in indoor fires, this study develops a coupled fire-evacuation simulation framework based on the lattice Boltzmann method (LBM) and the social force model (SFM). A supermarket scenario is used as the computational domain, and the text-based map is discretised into a numerical grid. A three-dimensional D3Q19-LBM scheme is adopted to describe low-Mach-number indoor ventilation flow, with temperature transport, semi-Lagrangian smoke advection and diffusion, buoyancy forcing, supply-air and exhaust boundaries, and moving-pedestrian obstacle feedback incorporated into the solver. Pedestrian motion is guided by an exit-distance potential field and accounts for interactions between pedestrians, wall repulsion, local congestion correction and speed reduction induced by smoke and heat exposure. Visibility, temperature and toxic-dose indicators are further used to quantify the influence of the fire environment on evacuation safety. Evacuation time, exposure dose, hazardous-area ratio, smoke-exhaust efficiency and population outcomes are obtained for different initial population sizes. The proposed framework therefore represents fire-environment evolution and pedestrian response within a unified computational procedure and provides a reproducible numerical basis for indoor fire-evacuation safety assessment, ventilation and smoke-exhaust optimisation, and pedestrian risk analysis.