Monte Carlo tree search for mass timber building design optimization

Mass timber construction has gained significant traction in recent years due to its sustainability and lower energy demands. However, its broader adoption remains limited by higher material costs, compared to conventional construction materials. To address this challenge, this study introduces a Monte Carlo tree search (MCTS)-based optimization framework aimed at minimizing the material cost of single-story post–beam–panel mass timber frame designs under gravity loads. By formulating the design task as a Markov Decision process, the MCTS algorithm can systematically guide step-by-step design decisions toward cost-efficient outcomes while satisfying structural constraints. The methodology is tested on four design scenarios modeled after real building dimensions. Results show that MCTS is capable of finding near-optimal solutions within just 1000 iterations, significantly reducing the computational effort required by exhaustive brute-force search. These findings underscore the effectiveness of MCTS as a promising tool for structural optimization in mass timber construction.