Model-Based Systems Engineering (MBSE) has become a standard for managing the complexity of modern systems. However, MBSE models are susceptible to internal inconsistencies, particularly between behavioral diagrams. A common and critical inconsistency arises when the sequence of interactions defined in a SysML Sequence Diagram cannot be realized by the behavior specified in the corresponding SysML State Machine Diagrams. Detecting these semantic discrepancies manually is an error-prone and time-consuming process that can allow design flaws to persist into later stages of the system lifecycle, significantly increasing costs and risks. This dissertation proposes an automated method to improve MBSE model quality by detecting these semantic inconsistencies by developing a dynamic consistency check approach that simulates the execution of State Machine Diagrams using the message sequence from a corresponding Sequence Diagram as a trace of input events. An inconsistency is identified if a state machine cannot process an event at any point in the sequence, indicating a conflict between the specified interaction and the designed behavior. The method was implemented for the open-source Capella MBSE tool to integrate seamlessly into the existing model validation, allowing engineers to run the consistency check on-demand. To validate the approach, the tool was applied to an case model of a rocket system containing intentionally and unintentionally introduced inconsistencies. The results demonstrate that the automated validation rule identified behavioral conflicts between the model's Exchange Scenarios (Sequence Diagrams) and Mode-State Machines.