<![CDATA[Background: Yogyakarta International Airport (YIA) is located on the southern coast of Java Island, which is prone to tsunamis due to tectonic plate subduction. The high passenger density in the terminal requires a fast, safe, and efficient evacuation system. However, evacuation route planning is generally still based on the shortest distance without considering passenger density, which has the potential to cause congestion during emergencies. Aims: This study aims to develop a tsunami evacuation route optimization model on the ground floor of YIA by simultaneously considering distance and passenger density factors through a weighted graph approach and the artificial bee colony metaheuristic algorithm. Method: Spatial data of the terminal layout is processed into a graph consisting of nodes and edges using QGIS. Edge weights are calculated from a combination of physical distance and passenger density. The optimization process is carried out using the ABC algorithm through the stages of population initialization, fitness evaluation, solution exploration, and selection of the best route to determine the minimum evacuation route from several starting points to the exit point. Result: The optimization results show that all ten initial evacuation points were efficiently allocated to six exit points with a minimum total weight. The resulting routes are not only shorter in terms of distance, but also avoid high-density areas, resulting in a more even distribution of passenger flow and reduced potential for congestion. Point EG was identified as the most optimal route based on a combination of distance and low density. Conclusion: The weighted graph approach based on artificial bee colony is effective in determining fast and adaptive tsunami evacuation routes in large-scale public facilities. This model has the potential to support more realistic disaster mitigation planning and can be applied to airports and other public infrastructure.]]>
