In my presentation, it has six sections. That is introduction, methodology, implementation, test analysis, and we have the conclusion in final.

In first section, we will introduce some background of this paper. In this page, we will use two examples to introduce the tsunami and flood disaster. The first example is the heavy rainfall in Japan in 2020 summer. The second example is the Tohoku earthquake and tsunami.

Because tsunami did not strike Ishinomaki in the last 100 years, a bridge with a height of 6m above Minxi level was chosen as the shelter of still deaths. And finally, this decision eventually caused 74 deaths of still deaths.

From these two examples, we can find that the high information plays an important role in tsunami and flood disaster. So an optimal evacuation model must consider the high information to locate a safe place for evacuation. In this page, we will talk about the 3D GIS.

Many 3D GIS applications nowadays are restricted to visualization purposes only. Because of lack of formalized and comprehensive mechanism for the management and analysis of future-based 3D geographic data. So the integration of 3D geospatial data and cross-domain information provides a comprehensive basis of future-based data framework.

As the cost of creating 3D data is often expensive, the development of 3DSDI is necessary to facilitate the successful sharing and use of coarse-domain 3D GIS data.

The major need of decision support system based on GIS is its capability to spatial analyze the given situation and evaluate different alternatives. The purpose system is designed as a decision support system. That allows users to improve real or simulated disaster scenarios and automatically respond with the visualized damage assessment information.

Next session, we will talk about methodology. The first we will talk about is the hierarchy of 3D building modeling.

In order to provide a physical operation module to meet the various demands of simulated scenarios, we designed a multi-level hierarchical framework of 3D building units. Such that the management, analysis and visualization can be subject to the chosen level, like building, story and household.

In this purpose, to ensure the correct linking between different level of 3D building units, every building unit is assigned a unique identifier. In this page, we will talk about identifier and linking to domain data.

Like the left side picture in this page, various domains of data are designed and linking by comma identifier. The 3D building framework survives as a core of 3D data with necessary geometry and essential attribute information.

The identifier information is used to linking to other available source of data or domain data. If the linking relationship can be created by comma identifiers or the mapping of domain-specific identifier, the scope of analysis can be fixed properly, expand without any limitation.

We have four components in smart decision support. The first component is to allow decision makers to design customized scenarios for disaster analysis. And the second component is about the visualization of 3D data.

And the third component is the variety of disaster loss analysis and visualization. And the final component is to provide decision makers a clear and objective reference for evaluating the trace from tsunami and floor disasters.

In section 3, we will talk about implementation. The test site in this purpose is the Wuzhe Township in Yilan County. It is one of area in Taiwan that receives the high impact in the simulated

tsunami potential analysis by the National Science and Technology Center for disaster reduction in Taiwan. The health of the area in Wuzhe Township has the tsunami flooding potential and the flood depth of one quarter of the area will reach 1 meter.

In this page, we will talk about the data preprocessing. In this purpose, four types of geospatial data and two types of non-geospatial data are used. As told in section 2, identifiers are used to link hierarchy 3D building, disaster potential, infrastructure and population data in our purpose.

In here, infrastructure data is used for showing the building is used for shelter, school, nursing home or not.

In section 4, we will talk about the text analysis. First, we will talk about the system operational mechanism. At first, the decision makers can select cross-domain data they wish to link for data preprocessing.

Then, decision makers can simulate disaster scenarios by specific disaster parameters like flooding depth. Based on the specific scenarios, the system can automatically generate the disaster assessment outcomes.

Then, the system will calculate and analyze the disaster loss index such as the trap population and affect buildings, stories and households. An outcome can represent on the 3D building frameworks for the decision makers to make decisions.

And finally, the purpose systems can be streamed on web3d.js platform. There are total five cases in our purpose.

The case 1 is the analysis and real representation of 3D over 4E. Like the right side picture in this page, the 3D representation provides a more realistic and intuitive representation for assessing the impacts of 4-lander 2D.

After analysis, buildings, stories or households which is impacted by the tsunami or falls are represented with red color to indicate their emergency state like the left side picture in this page.

The case 2 is the visible disaster index. We designed the user-customized index of disaster loss of population and damages of properties in this system. And like the picture in this page, total 18 index and 5 charts are designed and grouped into 3 categories.

And the high-risk query index can be used to highlight those buildings, stories or households whose number of variability populations exceeds the threshold values to make spatial evacuation decisions.

And the case 3 is about the multi-dimensional visualization with different applications and level of hierarchy.

There are two examples to show the physical visualizations of the purpose systems according to the chosen applications and level of building units. The example 1 is about simulating the overflowing situation and predict the loss at different times during the tsunami strikes.

Unlike other disasters, time factor plays an important role in tsunami disasters. And example 2 is about the visualization according to the quantitative measure of chosen topics.

Like the lower picture in this page, two buildings have different maximum capacity for evacuation. The dark color represents the higher capacity of the building and the lighter color for lower capacity.

The case 4 is the decision-making of available evacuation place. These systems can present the evacuation capacity in both the level of building and story.

So decision-makers can select the appropriate levels for making the evacuation decisions. For example, the building level can be used for regional evacuation and story level can be used for more precise arrangement of vertical evacuation.

The final case is the evacuation simulation. There are two different concepts in the purpose framework of evacuation. One is the people concept, using network analysis to determine the shelter and

evacuation routines with minimal working time according to the current location of people. And the other concept is based on shelter, using the concept of device area to analyze. In the final section, we will talk about the conclusion in this paper.

Many 3D dress applications nowadays are still restricted to visualization purpose. By proposing a web-based 3D dress system, we demonstrate how the disaster management can be improved by additionally considering the 3D characteristics for the traits of tsunami and floor disasters.

So we have the three major contributions in this paper. First is the hierarchy 3D building framework. This framework can provide a consistent geospatial reference for modeling various levels of building units and

enable simulation analysis and visualization of disaster impacts according to the levels of building units chosen. The second is the identifier system across domain data. We decide the common identifier system for domain data linkage and exchange to facilitate the development of 3D SDI.

And the final component is the integration of web technology and 3D dress. Decision makers can design, analyze scenarios, and alternatively impact the analysis outcomes for assessing the damage and making evacuation decisions.

And this page shows all of the references we use in this paper. And this is my presentation. Thanks for your listening.