Sambuaga, R.D. and Lee, H.S., 2021. Optimized evacuation plan and decision support system development with agent-based modelling and GIS analysis for tsunami events in Pandeglang, Banten, Indonesia. In: Lee, J.L.; Suh, K.-S.; Lee, B.; Shin, S., and Lee, J. (eds.), Crisis and Integrated Management for Coastal and Marine Safety. Journal of Coastal Research, Special Issue No. 114, pp. 509–513. Coconut Creek (Florida), ISSN 0749-0208.

On December 22, 2018, a tsunami caused by the eruption of the volcanic mountain and partial collapse of the Mt. Anak Krakatoa (The child of Krakatoa) volcano occurred in the Sunda Strait (a strait between Java and Sumatera island), Indonesia. The tsunami in the Sunda Strait resulted in 430 casualties (January 7th, 2019). The aim of this study is to provide optimum evacuation shelters by reducing evacuation time and casualty number caused by tsunami events. Early evacuation is the most significant measure in reducing casualties in tsunami disaster. Panimbang sub-district is chosen as a study area in this research, which it has the highest casualties according to Pandeglang Regional Disaster Management Agency (BPBD Pandeglang). Digital Elevation Model (DEM), information dataset (road network, hospital, worship facilities, schools, population distribution, etc.), and casualties' information are used in the analysis. This study conducts two parts; shelter analysis and agent-based modeling. The service area evaluation of proposed shelter is conducted using network analyst tool in ArcGIS. The important variable in the shelter analysis is the nearest accessibility based on the road network from residential area to proposed shelters. Public facilities such as mosque, school, health center, etc. are used as proposed shelters in the analysis. Ages group, walking speed, travel mode (walk or driving car), etc. are used as parameters and scenarios in the agent-based model. It affects total evacuee in this model.