Evaluating Industrial Activities and Its Social/Economic Impact in a Developing Country with Remotely Sensed Data: A Case Study of Eastern Economic Corridor, Thailand

Abstract

The industrial park is one of economic development tools that is based on integrating different facilities for producing goods and services, including manufacturers, infrastructures, and logistical supports (i.e., roads, utilities, schools, and medical services). However, the assessment of industrial development and its impacts is challenging, and the study at the local level is still limited in developing countries because of the lack of resources and limited budgets. This thesis examines the feasibility of using remotely sensed data and various methods to assess industry development and its social/economic impacts locally using Thailand's Eastern Economic Corridor (EEC) as the case study area.

This dissertation includes four inter-related parts of studies. In the first part, a rapid extraction method was employed in Google Earth Engine (GEE) to identify the built-up areas and understand the development status. Due to the incomplete household surveys in Thailand, in the second part, the missing household income was imputed using the multiple imputation method based on K-NN function with demographic and satellite-derived data. In the third part, the local electricity consumption was estimated using a decision tree regression model based on the variables from the second part to indicate industrial and economic activities. Industrial growth and its impacts were investigated in the fourth part using time-series of nighttime light data, detecting space-time patterns through an emerging hotspot analysis tool. This phase aims to identify the development patterns and their spatial spillover effect on the multidimensional poverty index of surrounding neighborhoods. The multi-temporal nighttime lights data illustrated growth patterns and were significantly correlated to GDP (Gross Domestic Product). Industrial development mainly helps poverty reduction. However, about 15% of neighborhoods still showed an increase in poverty with a low quality of life due to population concentration. This research demonstrates that the remotely sensed data can help study industrial activities and their impacts at the local level and better understand the spatial association of human activities and their impact on communities. The results will benefit national and local governments to better understand the consequences of implementing industrial development policies in less developed countries.