Session 3C: Big Data Analysis (Auditorium 3)
Moderator: Rana Hasan
Using big data to assess the impact of COVID-19 pandemic on tourism in Thailand analysis
Madhavi Pundit, Daniel Boller; Yothin Jinjarak; Yoo Ri Kim (University of Surrey), Anyu Liu (The Hong Kong Polytechnic University), Paolo Magnata, Elaine S. Tan, Priscille C. Villanueva
Thailand's economy is highly reliant on the tourist industry, making it vulnerable to disruptions arising from within and outside the country. Timely and granular data are, however, not available to provide detailed analyses of such disruptions. To assess the repercussions of COVID-19 pandemic on the Thai tourism industry, and more broadly, on Thailand's economy, the study will use mobile app data to model visitor flow patterns at significant points-of-interest before and during the pandemic; as well as to use those algorithms to quantify the impact of the pandemic and related containment measures on Thailand.
JEL Code/s: Z3, C01
Ship location data as real-time indicator in economic disruptions
Mahinthan Mariasingham, Elaine Tan; Cherryl Chico; Amna Gul, Ed Kieran Reyes
We assess the utility of Automated Integrated System (AIS) data, or location tracking data transmitted by ships every few seconds, in estimating real-time impact of recent disruptions, such as the Sri Lanka economic crisis, Russian invasion of Ukraine, and Tonga’s volcanic eruption. As a common challenge in the use of AIS data is identifying seaport boundaries, we develop a standardized method of identifying Areas of Interests (AOI) using Uber’s H3 index as basis for density-based cluster parameters. From these AOIs, we develop indicators based on maritime statistics and derived port calls to study the effects of these disruptions.
JEL Code/s: C8
Using satellite imagery and artificial intelligence to detect road quality
Ron Lester Durante and Arturo Martinez Jr.
Roads are a crucial part of every country’s physical infrastructure since they support the transportation of people, goods, and services, among others. However, the operation and maintenance of existing roads are typically underfunded resulting in a scarcity of quality information. Using satellite imagery and artificial intelligence, the study aims to develop an alternative approach for collecting road-quality data that is efficient and cost-effective. Based on the results, the algorithm's accuracy rate of up to 75% shows potential for preliminary identification or assessment of road quality.
JEL Code/s: O18, R42
Implications of transportation improvements and land use policy: The case of Bengaluru
Liming Chen, Rana Hasan, Yi Jiang and Andrii Parkhomenko (University of Southern California)
We study the effect of an ADB-financed metro network on economic welfare under different land use policies by building a quantitative spatial model of Bengaluru with multiple transportation models and heterogeneous workers. The model is calibrated with innovative big data. The height gradient patterns suggest that heights at city center are restricted by regulations. Simulations show that the metro system can bring significant welfare gains to all workers, especially for the high-skilled. The size of welfare gains depends crucially on the relaxation of land use regulations that will increase the supply of floorspace at city center in response to changes in demand.
JEL Code/s: R14, R31, R41, R42, R52