To speed up the automotive design and engineering process, it has become desirable to reduce the number of physical tests and prototypes . Part of the design process that is already performed virtually is the simulation of auto structural performance including structural, acoustic, and crash modeling. Each of these models still requires detailed inputs to properly characterize the strength of individual joints. Typically, this consists of a few standard joint pull tests to capture the primary failure behaviors: lap shear, coach peel, and cross tension tests or test variants including the KS2 test. These tests are commonly completed for each combination of material and thickness. Moreover, variations in material properties and process variables are inevitable in body structure joining. Examples of process variables may include electrode alignment, electrode force, electrode cap condition (due to wear), welding current and coating variations. Evaluating a matrix of material and process variation via physical testing becomes too time consuming and economically unfeasible. To greatly reduce the number of physical resistance spot weld (RSW) pull tests, it is necessary to have a reliable, repeatable, and predictive modeling and simulation approach. The first part of this work further developed an integrated approach that efficiently integrates together the RSW process simulation and pull test process simulation to yield accurate joint strength predictions. Validating the general workflow from RSW process simulation to tensile test of DP980 and GEN3-980 steels, many details were elucidated and a sensitivity analysis was also performed. The validation was done in terms of predicting nugget size given a welding schedule as input and using the resulting heat affected zone (HAZ) temperature profile to map experimentally obtained stress-strain curves. By combining the mapped stress-strain data with the Johnson-Cook damage criterion, the predicted force-displacement curve showed good agreement with experimental data from other A/SP characterization projects. During the execution, current Simufact solver capabilities were extended to address a handful of the identified gaps in the proposed workflow. One of the major contributions in predicting nugget size and correctly capturing the HAZ temperature distribution was an improved contact resistance model capable of addressing the complex interplay between contact pressure, local temperature, coating resistivity, electrode shape and bonding between the two sheets. Moreover, it was found that considering HAZ mechanical properties is essential to the accuracy of predicted peak load.