University of Twente Student Theses
The relation between street pattern and traffic congestion, an investigation through machine learning approach
Wismadi, M.T. (2022) The relation between street pattern and traffic congestion, an investigation through machine learning approach.
|
PDF
4MB |
| Abstract: | Traffic congestion is a problem that many cities face. To address this problem, policymakers frequently implement traffic interventions such as changing street directions or street access without building new roads. Such an approach may result in an unexpected distribution of traffic flows and the emergence of new congestion. This study attempts to understand this relationship between the street and traffic patterns. It is hypothesized that changes in traffic patterns, as well as the emergence of new congestion, are caused by traffic modifications which ultimately lead to changes in intricate street patterns. To test this hypothesis, a machine learning modelling approach is proposed in parallel to explore the potential of machine learning models in capturing the relationship between street morphology and traffic patterns. Using Barcelona and its Superblock policy as a case study, first, a traffic mapping comparison between before and after ‘traffic modification’ is conducted. Subsequently, change in street indices and the classification of street indices into street pattern types are explored. Finally, machine-learning traffic modelling is performed to compare linear regression, decision tree, and random forest algorithms. After estimating these without distinguishing by street type, the algorithm's evaluation is repeated by differentiating between street pattern types. |
| Item Type: | Essay (Master) |
| Faculty: | ITC: Faculty of Geo-information Science and Earth Observation |
| Subject: | 38 earth sciences, 50 technical science in general, 54 computer science, 55 traffic technology, transport technology, 74 (human) geography, cartography, town and country planning, demography |
| Programme: | Geoinformation Science and Earth Observation MSc (75014) |
| Link to this item: | https://purl.utwente.nl/essays/91208 |
| Export this item as: | BibTeX EndNote HTML Citation Reference Manager |
Show download statistics for this publication
Show download statistics for this publicationRepository Staff Only: item control page