{"id":{"repo_id":"texas-state","oai_identifier":"oai:digital.library.txst.edu:10877/22937"},"canonical_url":"https://search.dev.ndltd.org/etd/texas-state/oai:digital.library.txst.edu:10877/22937","repository":{"repo_id":"texas-state","name":"Texas State University","base_url":"https://digital.library.txst.edu/server/oai/request"},"display":{"title":"Artificial Intelligence and Spatial Modeling to Estimate Traffic Volume Measures on Local Roadways","abstract":"This study explores the integration of artificial intelligence (AI) and spatial modeling techniques to estimate Annual Average Daily Traffic (AADT) on local roadways, which are often data-scarce yet crucial for transportation planning and infrastructure development. Traditional traffic monitoring methods, such as permanent traffic count stations and short-term manual counts, are cost-prohibitive and fail to capture the variability and complexity of traffic flow on low-volume roads. To address this gap, the research develops and compares two modeling frameworks: a non-spatial Random Forest (RF) model and an enhanced spatial RF model. Using the comprehensive Statewide Traffic Monitoring Program (STMP) and Smart Location Database (SLD) dataset from Texas that incorporates socioeconomic, land use, environmental, and transportation accessibility variables, the study applies advanced machine learning methods to capture nonlinear relationships and interaction effects. The spatial RF model, augmented with geospatial diagnostics and cross-validation, demonstrates superior predictive performance over both the non-spatial RF and conventional Geographically Weighted Regression (GWR) models. Key predictors influencing traffic volume include regional centrality, transit ridership, and employment-residential balance. The results reveal complex, context-dependent relationships, emphasizing the importance of spatial heterogeneity and urban form in shaping traffic demand. The findings contribute valuable insights to data-driven traffic estimation on local roads, with practical implications for sustainable transportation planning, emission control, and equitable infrastructure investments. The study concludes by identifying model limitations and proposing future directions for improving the integration of dynamic temporal data and enhancing the interpretability of AI-based traffic models.","abstract_html":"This study explores the integration of artificial intelligence (AI) and spatial modeling techniques to estimate Annual Average Daily Traffic (AADT) on local roadways, which are often data-scarce yet crucial for transportation planning and infrastructure development. Traditional traffic monitoring methods, such as permanent traffic count stations and short-term manual counts, are cost-prohibitive and fail to capture the variability and complexity of traffic flow on low-volume roads. To address this gap, the research develops and compares two modeling frameworks: a non-spatial Random Forest (RF) model and an enhanced spatial RF model. Using the comprehensive Statewide Traffic Monitoring Program (STMP) and Smart Location Database (SLD) dataset from Texas that incorporates socioeconomic, land use, environmental, and transportation accessibility variables, the study applies advanced machine learning methods to capture nonlinear relationships and interaction effects. The spatial RF model, augmented with geospatial diagnostics and cross-validation, demonstrates superior predictive performance over both the non-spatial RF and conventional Geographically Weighted Regression (GWR) models. Key predictors influencing traffic volume include regional centrality, transit ridership, and employment-residential balance. The results reveal complex, context-dependent relationships, emphasizing the importance of spatial heterogeneity and urban form in shaping traffic demand. The findings contribute valuable insights to data-driven traffic estimation on local roads, with practical implications for sustainable transportation planning, emission control, and equitable infrastructure investments. The study concludes by identifying model limitations and proposing future directions for improving the integration of dynamic temporal data and enhancing the interpretability of AI-based traffic models.","abstract_has_math":false,"creators":["Mimi, Mahmuda Sultana"],"institution":"Texas State University","degree_name":"Master of Science","degree_level":"Masters","degree_discipline":"Civil Engineering","degree_department":null,"school":null,"contributors":[],"advisors":["Das, Subasish"],"committee_chairs":[],"committee_members":["Dutta, Anandi","Yuan, Yihong"],"year":2025,"date_issued":"2025-08","date_published":"2025-08","updated_at":"2026-07-27T21:22:32Z","subjects":["AADT","spatial RF","spatial AI","local roads","traffic volume estimation","low traffic volume"],"languages":["en"],"rights":[],"rights_urls":[],"identifier_entries":[]},"links":{"outbound_url":"https://hdl.handle.net/10877/22937","outbound_label":"Handle","outbound_source":"dc:identifier.uri"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor.advisor","label":"Advisor","values":["Das, Subasish"]},{"key":"dc:contributor.committeemember","label":"Committee Member","values":["Dutta, Anandi","Yuan, Yihong"]},{"key":"dc:creator","label":"Author","values":["Mimi, Mahmuda Sultana"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date.accessioned","label":"Dc Date Accessioned","values":["2025-10-09T18:17:17Z"]},{"key":"dc:date.issued","label":"Date","values":["2025-08"]},{"key":"dc:type","label":"Dc Type","values":["Thesis"]},{"key":"thesis:degree_discipline","label":"Discipline","values":["Civil Engineering"]},{"key":"thesis:degree_level","label":"Degree Level","values":["Masters"]},{"key":"thesis:degree_name","label":"Degree Name","values":["Master of Science"]},{"key":"thesis:institution_name","label":"Thesis Institution Name","values":["Texas State University"]}]},{"id":"subjects_keywords","label":"Subjects and Keywords","entries":[{"key":"dc:subject","label":"Dc Subject","values":["AADT","spatial RF","spatial AI","local roads","traffic volume estimation","low traffic volume"]}]},{"id":"language_rights","label":"Language and Rights","entries":[{"key":"dc:language.iso","label":"Language (ISO)","values":["en"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier.uri","label":"Identifier URI","values":["https://hdl.handle.net/10877/22937"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description.abstract","label":"Abstract","values":["This study explores the integration of artificial intelligence (AI) and spatial modeling techniques to estimate Annual Average Daily Traffic (AADT) on local roadways, which are often data-scarce yet crucial for transportation planning and infrastructure development. Traditional traffic monitoring methods, such as permanent traffic count stations and short-term manual counts, are cost-prohibitive and fail to capture the variability and complexity of traffic flow on low-volume roads. To address this gap, the research develops and compares two modeling frameworks: a non-spatial Random Forest (RF) model and an enhanced spatial RF model. Using the comprehensive Statewide Traffic Monitoring Program (STMP) and Smart Location Database (SLD) dataset from Texas that incorporates socioeconomic, land use, environmental, and transportation accessibility variables, the study applies advanced machine learning methods to capture nonlinear relationships and interaction effects. The spatial RF model, augmented with geospatial diagnostics and cross-validation, demonstrates superior predictive performance over both the non-spatial RF and conventional Geographically Weighted Regression (GWR) models. Key predictors influencing traffic volume include regional centrality, transit ridership, and employment-residential balance. The results reveal complex, context-dependent relationships, emphasizing the importance of spatial heterogeneity and urban form in shaping traffic demand. The findings contribute valuable insights to data-driven traffic estimation on local roads, with practical implications for sustainable transportation planning, emission control, and equitable infrastructure investments. The study concludes by identifying model limitations and proposing future directions for improving the integration of dynamic temporal data and enhancing the interpretability of AI-based traffic models."]},{"key":"dc:format","label":"Dc Format","values":["Text"]},{"key":"dc:format.medium","label":"Dc Format Medium","values":["1 file (.pdf)"]},{"key":"dc:title","label":"Title","values":["Artificial Intelligence and Spatial Modeling to Estimate Traffic Volume Measures on Local Roadways"]}]}],"canonical_facts":{"dc:contributor.advisor":["Das, Subasish"],"dc:contributor.committeemember":["Dutta, Anandi","Yuan, Yihong"],"dc:creator":["Mimi, Mahmuda Sultana"],"dc:date.accessioned":["2025-10-09T18:17:17Z"],"dc:date.issued":["2025-08"],"dc:description.abstract":["This study explores the integration of artificial intelligence (AI) and spatial modeling techniques to estimate Annual Average Daily Traffic (AADT) on local roadways, which are often data-scarce yet crucial for transportation planning and infrastructure development. Traditional traffic monitoring methods, such as permanent traffic count stations and short-term manual counts, are cost-prohibitive and fail to capture the variability and complexity of traffic flow on low-volume roads. To address this gap, the research develops and compares two modeling frameworks: a non-spatial Random Forest (RF) model and an enhanced spatial RF model. Using the comprehensive Statewide Traffic Monitoring Program (STMP) and Smart Location Database (SLD) dataset from Texas that incorporates socioeconomic, land use, environmental, and transportation accessibility variables, the study applies advanced machine learning methods to capture nonlinear relationships and interaction effects. The spatial RF model, augmented with geospatial diagnostics and cross-validation, demonstrates superior predictive performance over both the non-spatial RF and conventional Geographically Weighted Regression (GWR) models. Key predictors influencing traffic volume include regional centrality, transit ridership, and employment-residential balance. The results reveal complex, context-dependent relationships, emphasizing the importance of spatial heterogeneity and urban form in shaping traffic demand. The findings contribute valuable insights to data-driven traffic estimation on local roads, with practical implications for sustainable transportation planning, emission control, and equitable infrastructure investments. The study concludes by identifying model limitations and proposing future directions for improving the integration of dynamic temporal data and enhancing the interpretability of AI-based traffic models."],"dc:format":["Text"],"dc:format.medium":["1 file (.pdf)"],"dc:identifier.uri":["https://hdl.handle.net/10877/22937"],"dc:language.iso":["en"],"dc:subject":["AADT","spatial RF","spatial AI","local roads","traffic volume estimation","low traffic volume"],"dc:title":["Artificial Intelligence and Spatial Modeling to Estimate Traffic Volume Measures on Local Roadways"],"dc:type":["Thesis"],"thesis:degree_discipline":["Civil Engineering"],"thesis:degree_level":["Masters"],"thesis:degree_name":["Master of Science"],"thesis:institution_name":["Texas State University"]},"updated_at":"2026-07-27T21:22:32Z"}