3D Surface Design of Bobsleigh and Luge Tracks via Performance-Based Optimization: Lessons Learned from Yanqing National Sliding Center

• A sparse representation of the continuous surface geometry of a sliding track • Simulation-based optimization of the design considering multiple conflicting objectives • Integrating computational modules within a commercial software for system implementation Sliding sports of bobsleigh, skeleton and luge involve high-speed and high-overload competition, where the geometric profile of the track surface strongly influences performance. Existing design approaches are largely dependent on experience and lack systematic, data-driven methodologies. This study proposes a performance-based parametric design framework for three-dimensional track surfaces. The framework includes three components: generator, simulator and optimizer. The generator encodes the track geometry with a limited set of parameters to generate design candidates. The simulator then evaluates the sliding performance on the candidate design using a sliding motion simulation model. The optimizer is a multi-objective optimization model, which refines key geometric parameters to obtain a balance between different goals including safety, competitiveness, and cost considerations. The generator, simulator, and optimizer are integrated within the Grasshopper platform, forming a unified design tool. Validation is performed using the Yanqing National Sliding Centre track in Beijing, confirming the accuracy of both track generation and motion simulation. The optimization process is demonstrated through a virtual track design for recreational purposes. Results confirm that the proposed method can effectively produce track designs that meet specified performance criteria, offering a robust tool for engineering applications.