Network analysis for determining the fastest evacuation routes in flood-prone areas of the Tuntang Watershed, Indonesia

Flooding in the Tuntang Watershed, Central Java Province, is a hydrological disaster that occurs regularly. Several factors, such as high rainfall, limited water catchment areas, and narrowing riverbanks, cause flooding. The amount of material and non-material losses can be minimized through an effective disaster mitigation process. One way is to determine and establish flood evacuation routes to shelters. This research aims to analyze the fastest and most effective evacuation route for residents using Geographic Information Systems. This research process includes analyzing flood-prone areas, ranging from highly vulnerable areas to areas that are not vulnerable. Furthermore, it analyzes population density, distribution of residential areas, natural physical conditions, and road conditions and identifies the location of the best evacuation center for residents affected by flooding. The method used is Sample Additive Weighting (SAW) to determine the best evacuation point and Network Analysis Fastest Route to determine the optimal evacuation route to the designated evacuation center. The results of this study include modeling a flood evacuation route with an estimated effective evacuation time of 30-45 minutes, consisting of 50 fastest alternative routes to appropriate shelters.


Introduction
Indonesia is a two-season country located on the equator between the two continents of Asia and Australia.The climate in Indonesia is influenced by the west and east monsoon, which makes Indonesia only have two seasons, the rainy and the dry seasons.The climate influenced by global warming conditions like today makes weather conditions erratic, where the dry season can be very long and the rainy season can be hefty.Extreme conditions during the rainy season can cause hydrological disasters such as flooding.This condition will affect the activities carried out by the community when faced with extreme weather conditions.One of the affected community activities is the agricultural sector.High rainfall makes slopes steeper, and higher water runoff makes groundwater reserves smaller, impacting agricultural land degradation [1] [2] [3] [4].Data from the National Disaster Management Agency (BNPB) shows that extreme weather conditions have an impact on increasing flood hazard zones in several areas, one of which is around the Tuntang watershed.Data shows that losses due to flooding around the Tuntang watershed occur periodically during the rainy season, causing economic losses of around 71,504 billion rupiah, as well as damage to buildings and public facilities estimated at 8,889 billion rupiah with an affected population of 1,973,079 people.Therefore, Tuntang watershed and its surroundings need a good disaster management system to minimize material losses and casualties.Educating the general public and students in schools about the importance of evacuation routes and appropriate and efficient evacuation sites will have a significant impact [4] [5] [6].
A proper evacuation location requires several parameters, such as distance from settlements, roads, river flow, and safety from flood hazards [4].Previous studies conducted by Coutinho-Rougers verified that the relationship between time and distance to the evacuation site on foot is influenced by several factors such as road width, type of paving, slope conditions, and pedestrian speed up to a specific limit [7] [8].According to Wiliam Nasmith's research in 1892, humans generally take 1 hour to cover a distance of 5000 meters and an extra hour every 600 meters of road height [9].In addition to the distance factor, the condition of the evacuation site must also be in good condition or proper condition by considering the minimum space required by each person.Each person (individual) requires at least 1.64 m 2 of space.The condition of the building and the area of land used for the location becomes a reference to determine a proper evacuation site by calculating the building area divided by the space requirements of each person to determine the maximum capacity and feasibility of the evacuation site conditions [10].This study aims to determine the feasible evacuation location and the fastest evacuation route in floodprone areas in the Tuntang watershed of Indonesia.

Data collection for each parameters
The data collected is data based on secondary sources consisting of six parameters, namely data on flood-prone zones of the Tuntang watershed, road network data, river network data, settlement data, land use data, and rainfall data, all of which are obtained from the geospatial information agency and the national disaster management agency of the Republic of Indonesia.

Processing of parameter data
The data is processed using two methods, namely the simple addictive weighting method to determine feasible evacuation locations and network analysis to determine the fastest evacuation route.

Technic for analyzing data
The data is analyzed using a spatial function and a buffer function that is analyzed using a weighting system with the Simple Addictive Weighting (SAW) method, analyzing the distance as well as the route with the fastest travel time using the network analysis method and comparison using data from the results of research that has been done before.The data is analyzed using spatial functions and buffer functions that are analyzed using a weighting system with the Simple Addictive Weighting (SAW) method, analyzing the distance and route with the fastest travel time using the network analysis method and comparison using data from research that has been done before.The weighting criteria table can be seen in Tables 1-7;   [11] The results of data processing using weighting with the simple addictive weighting method will be a reference for determining locations that are suitable to be used as evacuation sites for communities affected by flood disasters.This data is then overlaid with road network data and settlement center points, which are then analyzed using the network analysis method with the shortest path and fastest route functions to determine the closest distance with the fastest travel time to the evacuation location points.

Results
The results below are the processing results of the six parameters using the simple addictive weigthing method and network analysis with spatial analysis using buffer functions and fastest routes. .Shows the results of flood zone hazard index data processing, with the highest flood hazard zone in the downstream area in Grobogan and Demak regencies.The high flood hazard occurs due to the narrowing of the riparian area, the distance between the river and settlements within 100-250 meters, and the high level of rainfall.The proximity of settlements and rivers can cause changes in the river flow profile and increase stormwater runoff.Figure 1 (b).Shows the results of rainfall data processing, showing the level of rainfall in the Tuntang River basin is at a value of 249 mm/month for the lowest rainfall and the highest with a value of 359 mm/month.The rainfall value is included in the high rainfall category.According to, [12] the greater the rainfall is related, the steeper the slope, which increases water runoff, making it difficult for water to be absorbed by the soil, increasing the potential for flood hazards. .Shows data on the distance of the river in the most at risk of flooding, namely the downstream area of the river basin with an average distance of 100-250 meters where the distance in this range is prone to flooding [11].Riparian areas that change function to become settlements or builtup land cause changes in river flow profiles that can accelerate rainwater runoff, thereby increasing the danger of flooding.Figure 2 (b) shows data processing results on the distance of settlements in the Tuntang River Basin area within 0-10 meters.The close distance will facilitate evacuation if a flood disaster occurs, but the distance of settlements that are too close can worsen the rainwater drainage process [13][14].This condition causes the Tuntang River basin, especially in the downstream part, to have a high flood hazard zone. .Shows the results of data processing.Road distance has a good accessibility value because the entire area is connected to the road, and the distance is only 500 meters, so the evacuation process during a flood disaster will be more straightforward.Figure 3 (b) land use and land cover data based on these data, the determination of evacuation or evacuation location points is determined based on four categories, namely open land, government buildings/offices, educational facilities (schools), and worship facilities.The land cover and land use map shows that the land cover for the gerobogan and demak districts is dominated by rice paddy fields, which are the most vulnerable to water runoff [15].
Figure 4. (a).Shows the data processing results using the sample addictive weighting method.The results show three categories: feasible, quite feasible, and not feasible.The condition of the evacuation site is feasible almost throughout the watershed area, Tuntang, and can be used as a temporary evacuation site before evacuating to a more feasible/permanent evacuation site by using government buildings, schools, and worship facilities scattered in settlement centers.Permanent evacuation sites, or those in proper condition, are evacuation sites that are safe from flood hazard zones and can accommodate a large number of evacuees by following the minimum space standard of 1.64 m 2 per person [16].26 viable evacuation sites can be used as permanent evacuation sites, all of which are open land.

Conclusions
From this study, it can be concluded From this research, it can be concluded that there are two categories of evacuation location points along the Tuntang River Watershed, namely not feasible, feasible enough, and feasible.For the category of feasible enough is spread almost throughout the watershed area, and feasible there are 26 points spread from upstream to downstream of Tuntang watershed.There are 50 alternative routes to the fastest feasible harvester evacuation site with an estimated effective evacuation time of 30-45 minutes.

Figure 1 .Figure 1
Figure 1.(a) Flood hazard map of the Tuntang watershed; (b) Road distance map of the Tuntang watershed

Figure 2 .Figure 2
Figure 2. (a).River distance map of the Tutang watershed; (b) Settlement distance map of the Tutang Watershed

Figure 4 .
Figure 3. (a).Shows the results of data processing.Road distance has a good accessibility value because the entire area is connected to the road, and the distance is only 500 meters, so the evacuation process during a flood disaster will be more straightforward.Figure3(b) land use and land cover data based on these data, the determination of evacuation or evacuation location points is determined based on four categories, namely open land, government buildings/offices, educational facilities (schools), and worship facilities.The land cover and land use map shows that the land cover for the gerobogan and demak districts is dominated by rice paddy fields, which are the most vulnerable to water runoff[15].Figure4.(a).Shows the data processing results using the sample addictive weighting method.The results show three categories: feasible, quite feasible, and not feasible.The condition of the evacuation site is feasible almost throughout the watershed area, Tuntang, and can be used as a temporary evacuation site before evacuating to a more feasible/permanent evacuation site by using government buildings, schools, and worship facilities scattered in settlement centers.Permanent evacuation sites, or those in proper condition, are evacuation sites that are safe from flood hazard zones and can accommodate a large number of evacuees by following the minimum space standard of 1.64 m 2 per person[16].26 viable evacuation sites can be used as permanent evacuation sites, all of which are open land.Figure4.(b) The results of data processing show that for the fastest evacuation route using the network analysis method obtained 50 alternative fastest routes to get to a feasible evacuation location with an estimated effective evacuation time of 30-45 minutes.

Table 3 .
River distance parameter

Table 4 .
Settlement distance parameter

Table 5 .
Road distance parameter

Table 6 .
Land use and Land cover parameter

Table 7 .
Evacuation site eligibility parameters