Mixture DOE Optimization of Playing Car Racing Video Game
Mason Chen
Stanford University
This paper will apply "STEAMS" methodology on gaming analytics. In the 21st century, the vast majority of youths are playing video games for too long (according to some studies an average of 13 hours/week). Parents do not want their children to play video games as they think it has a negative effect on their children. Chosen based on its wide applications of physics, Hill Climb Racing is the video game used in this project. Technology is applied to increase the transportation safety. Based on the engineering failure mode analysis and return of investment (ROI), a systematic car upgrading system was developed through statistical modeling to optimize the car performance. Several physics applications such as kinematics, energy/power, momentum, friction, circular motion, and gravity were applied on the car racing mechanisms. The AI clustering analysis grouped similar field stages (based on challenges, terrain, physics, and etc.) and cars. This increases cost efficiency and helps avoid wasted investment on upgrading multiple cars with similar functions. The statistical Mixture DOE optimizes the upgrading strategy based on the limited budget while helping enhance ROI.