Filtering by
- All Subjects: GIS
- Creators: Kuby, Michael
- Creators: Kuby, Michael J
Concerns about Peak Oil, political instability in the Middle East, health hazards, and greenhouse gas emissions of fossil fuels have stimulated interests in alternative fuels such as biofuels, natural gas, electricity, and hydrogen. Alternative fuels are expected to play an important role in a transition to a sustainable transportation system. One of the major barriers to the success of alternative-fuel vehicles (AFV) is the lack of infrastructure for producing, distributing, and delivering alternative fuels. Efficient methods that locate alternative-fuel refueling stations are essential in accelerating the advent of a new energy economy. The objectives of this research are to develop a location model and a Spatial Decision Support System (SDSS) that aims to support the decision of developing initial alternative-fuel stations. The main focus of this research is the development of a location model for siting alt-fuel refueling stations considering not only the limited driving range of AFVs but also the necessary deviations that drivers are likely to make from their shortest paths in order to refuel their AFVs when the refueling station network is sparse. To add reality and applicability of the model, the research is extended to include the development of efficient heuristic algorithms, the development of a method to incorporate AFV demand estimates into OD flow volumes, and the development of a prototype SDSS. The model and methods are tested on real-world road network data from state of Florida. The Deviation-Flow Refueling Location Model (DFRLM) locates facilities to maximize the total flows refueled on deviation paths. The flow volume is assumed to be decreasing as the deviation increases. Test results indicate that the specification of the maximum allowable deviation and specific deviation penalty functional form do have a measurable effect on the optimal locations of facilities and objective function values as well. The heuristics (greedy-adding and greedy-adding with substitution) developed here have been identified efficient in solving the DFRLM while AFV demand has a minor effect on the optimal facility locations. The prototype SDSS identifies strategic station locations by providing flexibility in combining various AFV demand scenarios. This research contributes to the literature by enhancing flow-based location models for locating alternative-fuel stations in four dimensions: (1) drivers' deviations from their shortest paths, (2) efficient solution approaches for the deviation problem, (3) incorporation of geographically uneven alt-fuel vehicle demand estimates into path-based origin-destination flow data, and (4) integration into an SDSS to help decision makers by providing solutions and insights into developing alt-fuel stations.
Universities host a large, young and diverse population that commutes to the same location every day, which makes them ideally suited for public transportation ridership. However, at many universities in the US, this potential for high levels of transit ridership is not being maximized. This research aims to identify the areas where Valley Metro’s public transit service to ASU’s Tempe campus is over- and under-performing in comparison with the overall public transportation service to the entire Phoenix metro area. The hypothesis states that proximity to campus and the convenience of using public transportation would be the two main factors in determining the success of an area’s public transportation service. ASU’s Parking & Transit Services provided confidential data with the addresses of all the students and employees who purchased a parking pass, transit pass and bike registration. With these data, the public transportation mode share for commuters to ASU in each census block group was calculated and compared to the mode share for the general public, which was based on US Census data. The difference between the public transit mode shares of ASU pass holders vs. commuting by the general public was then computed and analyzed to identify areas as hot and cold spots. These heat maps are then compared to the hypothesized factors of proximity to campus and the convenience of public transportation in terms of the light rail line, park-and-ride lots, and number of transfers needed to connect to campus. The transfers were estimated using origin and destination survey data provided by Valley Metro. Results show that the convenience of public transportation was a driving factor in explaining where the transit mode share to ASU is higher than that of the general public, whereas the proximity to campus had little impact on the areas with high ASU-specific transit mode shares. There is an absence of hot spots directly around the campus which is explained by the combination of both high transit share for the non-ASU population and the large share of ASU students and employees using active transportation and free circulator buses this close to campus. These findings are significant specifically to ASU because the university can learn where the transit service is performing well and where it is underperforming. Using these findings, ASU PTS can adjust its pricing, policies, services and infrastructure and work with Valley Metro and the City of Tempe to improve the ridership for both students and employees. Future research can compare more factors to further interpret what leads to success for transit service to university campuses.