Enhancing equity in public transportation using Geographic Information Systems and spatial optimization.

摘要

Public transportation is a vital part of urban living. Public transportation agencies receive substantial governmental subsidies due to the significance of their impact on society. For instance, public transportation services help reduce road congestion, oil consumption and air pollution, and they serve people who need to travel throughout urban environments at the same time do not have access to private vehicles. The latter aspect is an important matter of social justice. Therefore, it is necessary to discuss how the issue of equity is addressed in public transportation. It is important to understand why the interest in equity in transport is growing, why public transportation should favor the transport disadvantaged, and why analyses of equity measurement and improvement are needed. Measuring the level of access to public transportation among the transport disadvantaged provides a theoretical basis for analyzing potential improvements in access by adjusting public transportation facility locations. This research will focus on examining what equity issues in transport are present, how they are implemented in current public transportation systems, and their limitations. Another aspect of this research is modeling approaches used in establishing public transportation infrastructure and systems. Using GIS and spatial optimization models, the level of access to public transportation in terms of equity will be evaluated and improvement of the level of access will be attempted by offering new service stop locations. To this end, the Polygon Intersection with Network Point Set (PINPS) method is developed to locate potential facilities along a continuous road network ensuring complete coverage of polygon demand objects. Using the Maximal Covering Location Problem (MCLP), the optimal locations of potential facilities to cover equity favoring origin- and destination-based demand are identified. This research finally provides a set of optimal service stop locations maximizing coverage of origin- and destination-based demand simultaneously through implementation of a bi-objective model, applied to the City of Hilliard, Ohio. In addition the capabilities of coverage models utilizing PINPS and randomly obtained potential facilities are compared. Also, the use of models that represent polygon-based demand versus those depending on only typical point-represented demand will be discussed.