• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.11 no.12
• /
• pp.4795-4802
• /
• 2010

In order to solve the traffic problem in jeju, we must reduce demand for car travel. In addition, demand for passenger travel by public transport policy is needed for conversion. And to improve the quality of public transport services are desperately needed. The purpose of this study, personal characteristics of the trip traveler and the relationship between transportation choice, and personal effectiveness as a factor in travel costs and travel time on the impact of transportation choices will investigate. Restructure its public transportation routes, when to switch to buses to car traffic on the data as a basis for the factors that may be. In addition to improving the quality of public transport services is expected to be able to contribute. To study the performance of the May 2010 survey was conducted. And multinominal logit model were conducted. According to the analysis, People who own homes and families with more than 5 people are likely to use cars. If a prolonged travel time is likely to use buses. However, increasing the cost of travel increases the likelihood that the car is available.

• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.19 no.3
• /
• pp.28-37
• /
• 2020

This study developed a deep learning model that predicts rental demand for public bicycles. For this, public bicycle rental data, weather data, and subway usage data were collected. After building an exponential smoothing model, ARIMA model and LSTM-based deep learning model, forecasting errors were compared and evaluated using MSE and MAE evaluation indicators. Based on the analysis results, MSE 348.74 and MAE 14.15 were calculated using the exponential smoothing model. The ARIMA model produced MSE 170.10 and MAE 9.30 values. In addition, MSE 120.22 and MAE 6.76 values were calculated using the deep learning model. Compared to the value of the exponential smoothing model, the MSE of the ARIMA model decreased by 51% and the MAE by 34%. In addition, the MSE of the deep learning model decreased by 66% and the MAE by 52%, which was found to have the least error in the deep learning model. These results show that the prediction error in public bicycle rental demand forecasting can be greatly reduced by applying the deep learning model.

• Journal of Korean Society of Transportation
• /
• v.29 no.5
• /
• pp.43-54
• /
• 2011

After implementing "Semi-public management system of intra-city bus", the burden of financial aid for unprofitable routes is on the increase in Busan metro city. It becomes a heavy burden on the local finance, which needs to be resolved for improving the intra-bus system. The rainfall is one of the factors influencing the demands for intra-bus in urban transportation. Motivated by this fact, this study investigates the impact of rainfall on the intra-city bus demand. Actual bus users are surveyed on their patterns and recognition of using the bus according to the amount of rainfall. A rainfall forecast model using ordered probit model is presented, and the elasticity of the intra-city bus utilization to the amount of rainfall is also analyzed. The resulting findings could be applied to promote the use of intra-city buses and also be utilized as basic data for other studies to improve the intra-city bus system.

• Journal of Korean Society of Transportation
• /
• v.31 no.2
• /
• pp.33-44
• /
• 2013

The land use-transportation models typically have complicated model structure that is good for empirical execution but bad for theoretical probe. This complexity makes it very difficult to derive the first-order conditions for system optimization in tractable forms. Yu and Rhee (2011) and Rhee (2012) show how to simplify the derivative of the model's objective function with respect to policy variables in the computable general equilibrium model of land use and transportation. However, the travel demand in their model was fixed. This drawback fundamentally limits the applicability of their methodology in the planning field. We relax this restriction. Once this is done, we can employ the methodology developed in analyzing the impacts of various types of policy instruments in the models where land market is treated endogenously and transportation network is embedded.

• Spatial Information Research
• /
• v.19 no.4
• /
• pp.11-20
• /
• 2011

In accordance with low carbon and green growth paradigm, a subway is one of major public transit systems for resolving traffic congestion and decreasing traffic accidents. In addition, as subway networks expand, passengers' travel pattern in the subway network change and consequently affect the urban structure. Generally, new subway route has been planned and developed, mainly considering a travel demand forecast. However, it is desired to conduct an empirical analysis on the forecast model regarding change of travel accessibility and passenger demand pattern according to new subway line. Therefore, in this paper, an alternative method, developed based upon a spatial syntax model, is proposed for evaluating new subway route in terms of passenger's mobility and network accessibility. In a case study, we constructed subway network data, mainly targeting the no 9 subway line opened in 2009. With an axial-map analysis, we calculated spatial characteristics to describe topological movement interface. We then analyzed actual modal shift and change on demand of passengers through the number of subway passenger between subway stations and the number of passenger according to comparative bus line from Smart Card to validate suggested methods. Results show that the proposed method provides quantitative means of visualizing passenger flow in subway route planning and of analyzing the time-space characteristics of network. Also, it is expected that the proposed method can be utilized for predicting a passengers' pattern and its impact on public transportation.

• Journal of Korean Society of Transportation
• /
• v.34 no.6
• /
• pp.510-522
• /
• 2016

In studies involving public transport, social welfare improvement is simply explained by the increase in public transport demand. However, the increase in the demand for public transport is mostly observed by the change in the frequency of public transport service, and in-vehicle crowding in public transport has not been an object of concern. This study examines and tries to reveal the cause of the changes of the social welfare and in-vehicle crowding of the changing public transport from imposing congestion pricing. We observe that congestion pricing increases in-vehicle crowding in public transport. This predictable phenomenon is more exacerbated in case of not operating bus-only lane. It should be noted that in-vehicle crowding is more increased in suburban, but in First-best toll system it tends to get worse less than it in other congestion pricing systems. We identify that the change of in-vehicle crowding is affected by the change of proximity of the housing to workplace, the number of commuting trips, and unpredictable distortion effect of the congestion charge.

• Journal of Korean Society of Transportation
• /
• v.17 no.5
• /
• pp.7-18
• /
• 1999

This Paper focuses on analyzing the influences of Teleservice on travel behavior and trip demand based on the individual behavior model. The first step is to classify individuals into different Person groups who will follow similar behavior Patterns in terms of travel and communication. And then, the effect of Teleservice on trip demand is estimated using hypothetical scenario. The results on the ability of each person group to adapt themselves to Teleservice show that full-time housekeepers and senior citizens are more likely to be alienated. It is also found that transition probability to Teleservice is high for activities in bank or public office where simple forms of information such as data or text are exchanged. On that basis, it is estimated that in Seoul the savings on trip demand by Teleservice will be 7.6% of total daily trip generations.

• Journal of Korean Society of Transportation
• /
• v.19 no.4
• /
• pp.97-108
• /
• 2001

A dynamic transportation management should be applied specially in AM peak because AM peak is more critical than PM peak in traffic volume and demand. AM peak trip can be characterized by commuting and schooling. which have the high level of usage on public transportation, and constraint on arrival time. So transportation management applied in AM peak could deal with a mode choice and an arrival time constrain. Researchers were involved in the dynamic transportation assignment models for management of congested traffic network. But, there were no models which considered a mode choice and an arrival time constrain should be included in management of AM peak. So there are limits to use exist models to control and analyze AM peak traffic. In this study, it is proposed the combined dynamic transportation model, considering a mode choice and the start time selection with arrival time constrains, based on Ran and Boyce's model. The proposed model is verified the compatibility by applying to the newly designed time space expanded network. The result shows that proposed model consistent with dynamic user optimal travel pattern. From this we certificate the applicability of the proposed model to control and analyze AM peak traffic.

• Smart Media Journal
• /
• v.11 no.11
• /
• pp.49-62
• /
• 2022

Recently, with the development of Smart City Solutions such as Big data, AI, IoT, Autonomous driving, and Digital twins around the world, the proliferation of various smart devices and social media, and the record of the deeds that people have left everywhere, the construction of Smart Cities using the "Big Data" environment in which so much information and data is produced that it is impossible to gauge the scale is actively underway. The Purpose of this study is to construct an objective and systematic analysis Model based on Big Data to improve the transportation convenience of citizens and formulate efficient policies in Urban Information and Public Transportation accessibility in sustainable Smart Cities following the 4th Industrial Revolution. It is also to derive the methodology of developing a Big Data-Based public transport accessibility and policy management Platform using a sustainable Urban Public DB and a Private DB. To this end, Detailed Living Areas made a division and the accessibility of basic living amenities of Gwangju Metropolitan City, and the Public Transportation system based on Big Data were analyzed. As a result, it was Proposed to construct a Big Data-based Urban Information and Public Transportation accessibility Platform, such as 1) Using Big Data for public transportation network evaluation, 2) Supporting Transportation means/service decision-making based on Big Data, 3) Providing urban traffic network monitoring services, and 4) Analyzing parking demand sources and providing improvement measures.

• Journal of Korean Society of Transportation
• /
• v.28 no.3
• /
• pp.145-155
• /
• 2010

As the green transportation mode, revitalization of bike usage attracts remarkable public attention. For the acquirement of effective outcome, however, the concrete and close analysis about bike utilization characteristics should be arranged first. One result by MLTM(2009) is support this opinion; the bike mode share has been decreased whereas 9,170km of the bicycle path was improved(1995~2007). This study analyzed the bike mode share classified by trip types by using the 303,308 data of Household Travel Survey of Seoul Metropolitan Area, 2006. The highest mode share rate was induced by the institute attendee and Officetel resident as 3.75% and 3.13%, respectively. Also this study established the bike mode share estimation model of Seoul by logistic regression, and analyzed related factors and level of effectiveness related bike demand by calculation of odds ratio in terms of logistic regression coefficients. In conclusion, short trips, institutes district, parks, and Officetel residential area oriented policy should be effective on the revitalization of bike usage.