Please use this identifier to cite or link to this item:
http://hdl.handle.net/1942/29512
Title: | Macro-level safety analysis using crash prediction models: The case of Metro Manila | Authors: | Lamar, Kimjay | Advisors: | PIRDAVANI, Ali | Issue Date: | 2019 | Publisher: | UHasselt | Abstract: | The study aimed to create traffic analysis zone (TAZ)-level crash prediction models for Metro Manila. It utilized the data available from the latest transportation study for the capital region. Using these data does not require additional budget for collection; therefore, it could become attractive to policy makers. Data that came from the transportation study, MMUTIS (Metro Manila Urban Transportation Integration Study, the previous transport study) Update and Enhancement Project (MUCEP), are trip information according to purpose and mode, day population ratio, and TAZ shape files. Additional data came from the Philippine Statistics Authority (population, number of household), PhilGIS (road lengths according to road classification), and Google Earth historical map (land use information). In this thesis we investigated three different methods of regression: global negative binomial (NB) regression, clustered negative binomial regression, and geographically weighted Poisson regression (GWPR). Global NB model served as the base model for comparison. In the clustered NB models, prior clustering of the TAZs was performed based on their characteristics. Then a negative binomial regression is performed in each cluster, which led to 4 models with 4 sets of coefficients. The GWPR model allows the coefficients to be different in each TAZ, resulting to 225 sets of coefficients. | Notes: | Master of Transportation Sciences-Traffic Safety | Document URI: | http://hdl.handle.net/1942/29512 | Category: | T2 | Type: | Theses and Dissertations |
Appears in Collections: | Master theses |
Files in This Item:
File | Description | Size | Format | |
---|---|---|---|---|
3588a626-50ba-4080-8f38-a20fbb2de125.pdf | 3.79 MB | Adobe PDF | View/Open |
Page view(s)
68
checked on Oct 29, 2023
Download(s)
32
checked on Oct 29, 2023
Google ScholarTM
Check
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.