For structural components that operate at elevated temperatures, martensitic P91 steel is preferable. It is widely used in steam generators in the fossil-fired thermal and nuclear power generation sectors due to its creep endurance and corrosion resistance. Several creep laws, such as Monkman-Grant, Theta project, Wilshire and Sinh model, Omega technique, and the Larson Miller Parameter (LMP), have been developed over time to predict and failure of materials susceptible to the creep phenomenon. However, only the Omega Law and Larson-Miller Parameter are the only two methods approved in API 579-1. In this work, the creep test for welded and unwelded P91were conducted at temperatures of 600°C under stresses of 165, 175 MPa, and 190 MPa. In comparison to the welded specimens, the unwelded specimens displayed a continuously more substantial development of creep strain with time, resulting in a higher steady-state creep rate and a shorter rupture life. The increased magnitude of creep-rupture data has been observed to impact the dependability of creep life. Because of more significant changes in service temperature and stress conditions, the dependability of P91 steel has deteriorated, as well as an increase in creep life. The rupture life has been predicted using the LMP method, which utilizes the constant C parameter. At the same stress, the predicted creep life for weld material shows a higher value than that of the parent material, which is consistent with the experimental result.