An ideal theoretical model integrating four revolute joints is developed to capture the kinematic behaviour of a vehicle scissor door joint mechanism. Then, triaxial acceleration experiments are conducted to validate the effectiveness of the developed theoretical model. Furthermore, to improve dynamic responses of the mechanism, a novel discrete multi-objective optimization (DMO) method is proposed to address optimization problems where design variables cannot be parameterized. This method integrates the Taguchi method, grey relational analysis and a hybrid multi-objective decision-making approach, and iteratively updates the orthogonal array to perform optimization for handling design variables with multiple levels. Compared to the conventional non-dominated sorting genetic algorithm-II (NSGA-II), the developed DMO is capable of achieving the Pareto frontier with fewer evaluations of the objective function. The optimization results reveal that the optimized design for the electric and gas struts exhibits favourable dynamic responses of scissor door operation compared to the initial design. [ABSTRACT FROM AUTHOR]