UTILIZATION OF MULTI-SOURCE SATELLITE IMAGERY TO ESTIMATE THE IMPACT OF CIPALI TOLL ROAD DEVELOPMENT ON VILLAGE ECONOMY

• Main Author: Nateq Nouri, Mh.
• Format: Theses
• Language: Indonesia
• Subjects: Perencanaan wilayah
• Online Access: https://digilib.itb.ac.id/gdl/view/84410
• Institution: Institut Teknologi Bandung

Infrastructure development is a key element in driving economic growth, with the potential to increase production levels, create job opportunities, and reduce poverty rates. One of the priority projects is the construction of the Trans Java toll road, planned to connect Merak to Banyuwangi by 2024. This toll road is expected to have a positive impact on economic development in the regions it passes through and contribute to economic equalization to reduce disparities in underdeveloped areas. This research highlights the construction of the Cipali Toll Road as a case study, illustrating how toll road infrastructure influences mobility, distribution efficiency, and regional economic activity. The Cipali Toll Road, which is part of the Trans Java series, stretches for 116.75 km and crosses five regencies: Subang, Indramayu, Majalengka, Purwakarta, and Cirebon. The vital role of the Cipali Toll Road in connecting various regions on Java Island means that it has a significant impact on the surrounding economy. However, existing economic research rarely targets the local level, such as villages, because conventional national surveys require considerable time and effort. This study will examine and identify the spatial economic impacts of the Cipali Toll Road on the villages it passes through during the 2013-2023 period. The data used in this study includes the Gross Regional Domestic Product (GRDP) at constant prices, adjusted for inflation, as the dependent variable, and independent variables comprising satellite imagery data that produce variables such as NTL, NDVI, NDBI, NDWI, LST, and population density at the district level. The analysis methods include developing an economic estimation model using machine learning, then using the small area estimation technique to obtain total and per capita economic values at the village level with predictive variables NTL, NDVI, NDBI, NDWI, LST, and population density. Furthermore, Moran's I analysis and LISA cluster analysis are used to understand the characteristics of economic distribution before and after the construction of the Cipali Toll Road. The final analysis technique, difference in difference, is used to calculate the economic impact by comparing villages traversed by the toll road with control villages that are not, based on the spatial distance from the toll exit to the length of the toll road.iv The results of this series of analyses indicate that utilizing multi-source satellite imagery variables combined with population density produces the best model across all districts, specifically the Random Forest Regression (RFR) model, with a confidence range (R²) of 60-92%. Following this, small area estimation was conducted to obtain estimates for total and per capita economic values. The spatial distribution in the LISA clusters demonstrated an equalization of per capita economic conditions, as evidenced by a decrease in both High-High and Low-Low clusters. However, there was an increase in the Low-Low cluster for total economic output, indicating a growing isolation of villages classified as having low economic levels. The economic impact, calculated through difference-in-difference analysis, reveals that villages within a 3 km radius of toll exit gates experienced a statistically significant total economic impact