Predicting the part thermal history during the selective laser melting (SLM) process is critical to understand the influence of the process parameters to the part quality. Existing finite element based thermal analysis is mainly associated with simplifications in mesh configuration, heat source model, and domain size. The proposed work presents an efficient adaptive remeshing technique that enables part-scale SLM process simulations and helps reduce model size without sacrificing accuracy. The proposed work enables the part-scale simulation computationally efficient using existing commercial solvers. In this paper, the SLM process simulation for an entire part was developed considering different process parameters. The model predicts the influence of the process parameters on part thermal history, melt pool statistics, and lack-of-fusion porosity. The predicted results find an agreement with the experimental results in literature. Furthermore, the remeshing technique is demonstrated to be more computationally efficient than the existing element death and birth approach and also shows clear advantages compared with existing adaptive remeshing approaches.