A Machine Learning Approach to the Residential Relocation Distance of Households in the Seoul Metropolitan Region

Analysis of Inter-Relationships between Urban Decline and Urban Sprawl in City-Regions of South Korea

This paper identifies inter-relationships between the urban decline in core areas and urban sprawl in hinterlands using 50 city-regions of South Korea. We measured decline- and sprawl-related indicators and estimated a simultaneous equations model using Three-Stage Least Squares. The results show that population decline and employment decline have a different relationship with urban sprawl. While population decline has a negative impact on the urban sprawl in the density aspect, employment decline worsens the urban sprawl in the morphological aspect. Another result suggests that the difference is related to declining patterns of population and employment. Cities that are experiencing population decline in the core area are likely to lose population in their hinterlands as well. On the other hand, the employment decline in the core area shows a positive correlation with employment growth in hinterlands. The results imply that suburbanization of jobs and the inefficient land use exacerbate the urban sprawl in the morphological aspect. Thus, local governments should pay attention to migration patterns of employment and make multi-jurisdictional efforts. Furthermore, growth management and urban regeneration policies should go hand in hand to tackle this issue.

Development of the Road Pavement Deterioration Model Based on the Deep Learning Method

In Korea, data on pavement conditions, such as cracks, rutting depth, and the international roughness index, are obtained using automatic pavement condition investigation equipment, such as ARAN and KRISS, for the same sections of national highways annually to manage their pavement conditions. This study predicts the deterioration of road pavement by using monitoring data from the Korean National Highway Pavement Management System and a recurrent neural network algorithm. The constructed algorithm predicts the pavement condition index for each section of the road network for one year by learning from the time series data for the preceding 10 years. Because pavement type, traffic load, and environmental characteristics differed by section, the sequence lengths (SQL) necessary to optimize each section were also different. The results of minimizing the root-mean-square error, according to the SQL by section and pavement condition index, showed that the error was reduced by 58.3–68.2% with a SQL value of 1, while pavement deterioration in each section could be predicted with a high coefficient of determination of 0.71–0.87. The accurate prediction of maintenance timing for pavement in this study will help optimize the life cycle of road pavement by increasing its life expectancy and reducing its maintenance budget.

Factors Influencing Matching of Ride-Hailing Service Using Machine Learning Method

It is common to call a taxi by taxi-apps in Korea and it was believed that an app-taxi service would provide customers with more convenience. However, customers’ requests can often be denied, as taxi drivers can decide whether to take calls from customers or not. Therefore, studies on factors that determine whether taxi drivers refuse or accept calls from customers are needed. This study investigated why taxi drivers might refuse calls from customers and factors that influence the success of matching within the service. This study used origin-destination data in Seoul and Daejeon obtained from T-map Taxis, which was analyzed via a decision tree using machine learning. Cross-validation was also performed. Results showed that distance, socio-economic features, and land uses affected matching success rate. Furthermore, distance was the most important factor in both Seoul and Daejeon. The matching success rate in Seoul was lowest for trips shorter than the average at midnight. In Daejeon, the rate was lowest when the calls were made for trips either shorter or longer than the average distance. This study showed that the matching success for ride-hailing services can be differentiated particularly by the distance of the requested trip depending on the size of the city.