Intersecting Geostatistics with Transport Demand Modeling: a Bibliographic Survey
Transport planning depends on modeling variables, and because their collection usually requires high resources, they have limited sampling. However, since they are spatially dependent, the use of Geostatistics in transport demand modeling has proved to be especially convenient, as this tool allows o...
| Autores: | , |
|---|---|
| Formato: | artículo |
| Estado: | Versión publicada |
| Fecha de publicación: | 2020 |
| País: | Brasil |
| Recursos: | Universidade Federal de Uberlândia (UFU) |
| Repositorio: | Revista brasileira de cartografia - RBC (Online) |
| Idioma: | portugués inglés |
| OAI Identifier: | oai:ojs.www.seer.ufu.br:article/56467 |
| Acesso em linha: | https://seer.ufu.br/index.php/revistabrasileiracartografia/article/view/56467 |
| Access Level: | acceso abierto |
| Palavra-chave: | Demanda por transportes Geoestatística Mobilidade urbana Krigagem Suporte geográfico Transportation demand Geostatistics Urban mobility Kriging Geographic support |
| Resumo: | Transport planning depends on modeling variables, and because their collection usually requires high resources, they have limited sampling. However, since they are spatially dependent, the use of Geostatistics in transport demand modeling has proved to be especially convenient, as this tool allows obtaining estimates in non-sampled locations. In this context, the research line related to applying Geostatistics to travel demand forecasting takes place within the scope of three of the four steps of the traditional planning model (trip generation, modal choice and traffic assignment), covering studies that can be divided according to the support, or geographic scale adopted, and type of model used. Thus, in order to establish the state-of-the-art of this research line, the present study proposed surveying and discussing articles within the scope of traffic zones, regular areas, road segments, metro stations, bus stops, bus line segments and household/individual analysis, which used Simple, Ordinary, Indicator, Universal and Spatio-temporal Kriging geostatistical interpolation, in addition to Gaussian Sequential Simulation. The detailed analysis of the studies allowed identifying research gaps in the models’ validation stage, comparison with other spatial and non-spatial approaches, use of network distances, applying Universal Kriging (UK) to modal choice variables and the selection of predictors for UK. Special attention should be given to Sequential Gaussian Simulation and Spatio-temporal Kriging, models that could dictate the evolution of the research line in the coming years. |
|---|