Developing a decision-support tool for footbridge planning and design phases

Date

2025-1-8

Author

Eloğlu, Bade

Metadata

Show full item record

This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

Item Usage Stats

103
views

41
downloads

Cite This

A livable city is rooted in the quality of the pedestrian experience. Yet still, pedestrians remain one of the most vulnerable groups within built environments. With policies prioritizing motor vehicle accessibility, cities are growing increasingly centered around vehicular roads, and the spaces for pedestrians to safely and comfortably navigate around are narrowing down day by day. In such environments, footbridges, particularly on multi-lane roads with high-speed traffic, have become the most prevalent examples of the ad-hoc manner in crossing facility construction. However, imposing these structures into cities without considering site-specific constraints and pedestrian expectations further restricts pedestrian mobility. To address this challenge, this thesis proposes a decision-support tool comprising two complementary stages for footbridge planning and design phases. It employs a mixed-methods approach that integrates on-site observations and literature review findings with machine learning methods. In the first stage, a decision tree model was trained with shared built environment features of footbridge locations to predict the need for footbridges in a given location. However, constructing footbridges merely due to the need is transforming them into urban barriers. Thus, in the second stage, the key factors affecting footbridge usage were identified through a review of literature, later incorporated with the researcher’s observations on footbridge sites and mandatory design criteria. These findings were synthesized into a design guideline, translating pedestrian expectations into actionable design principles. In the final, using the guideline as the main source, the necessity and feasibility of constructing footbridges in the predicted locations were critically evaluated by the researcher.

Subject Keywords

Footbridge, Pedestrian Preferences, Prediction Model, Machine Learning, Design Guideline

URI

https://hdl.handle.net/11511/113488

Collections

Graduate School of Natural and Applied Sciences, Thesis