With rising global temperatures, heat-related illnesses pose a significant risk to tourists. This study proposes a methodology for generating tourist routes that balance sightseeing experience with heat safety. We first introduce a method for estimating tourist heat stress by adapting the ACGIH framework. Subsequently, we employ a multi-objective optimization algorithm, NSGA-II, to generate tourist routes. The algorithm maximizes a tourist route score derived from social media data and minimizes transit time, while satisfying a heat stress constraint. An experiment was conducted based on tourist spots in Okayama, Japan. The results show that on days with higher heat risk, the algorithm generates safer routes by reordering tourist spots or suggesting travel to different areas to reduce heat stress exposure.