Abstract Laser powder bed fusion (LPBF) is an additive manufacturing technique that utilizes laser-induced melting of specific regions within a powder layer to create complex parts. Achieving high-quality products in LPBF requires the optimization of process parameters based on the unique characteristics of the powder material. Since experimental optimisation can be both time-consuming and costly, we propose a computational model capable of simulating the particle micro-mechanics in LPBF, offering a more cost-effective solution. We have developed a novel thermal discrete particle and contact model that accurately captures the essential phenomena of melting, coalescence, and consolidation within LPBF. Our model assumes that solid particles partially melt under the influence of heat, subsequently coalesce, and form solid bonds during the cooling phase. The rate of coalescence is determined by the material’s surface tension and viscosity as it undergoes melting. To account for phase transitions, we employ an apparent heat capacity method. We first introduce our contact model and provide verification against analytical solutions for a two-particle system. We then demonstrate the efficacy of our model by applying it to a multi-particle example, successfully capturing the coalescence and consolidation behaviour observed in LPBF. The model has been implemented in the open-source code MercuryDPM. The current model is developed for polymer material, but it can be extended to metal and ceramic. Graphical Abstract