View Item 
      •   IPB Repository
      • Dissertations and Theses
      • Undergraduate Theses
      • UT - School of Data Science, Mathematic and Informatics
      • UT - Statistics and Data Sciences
      • View Item
      •   IPB Repository
      • Dissertations and Theses
      • Undergraduate Theses
      • UT - School of Data Science, Mathematic and Informatics
      • UT - Statistics and Data Sciences
      • View Item
      JavaScript is disabled for your browser. Some features of this site may not work without it.

      Kajian Analisis Pola Kemacetan Ruas Jalan Jakarta Selatan menggunakan Spatio Temporal Graph Convolutional Network

      Thumbnail
      View/Open
      Cover (588.6Kb)
      Fulltext (1.980Mb)
      Lampiran (1.235Mb)
      Date
      2025
      Author
      Taufiq, Fadly Mochammad
      Susetyo, Budi
      Alamudi, Aam
      Metadata
      Show full item record
      Abstract
      Penelitian ini bertujuan untuk memodelkan dan mengeksplorasi panjang kemacetan lalu lintas pada ruas jalan di Jakarta Selatan dengan pendekatan Spatio Temporal Graph Convolutional Network (ST-GCN). Model ini menggabungkan Graph Convolutional Network (GCN) untuk menangkap keterkaitan spasial dan Gated Recurrent Unit (GRU) untuk memahami dinamika temporal. Data kemacetan diperoleh dari Smart City Jakarta dan dipadukan dengan data spasial OpenStreetMap. Setelah konstruksi graf dan praproses data, model ST-GCN dilatih dan dibandingkan dengan model baseline GRU dan ST-GAT menggunakan metrik MAE dan SMAPE. Hasil menunjukkan bahwa ST-GCN memberikan prediksi yang cukup akurat, dengan SMAPE rata-rata sebesar 8,70% Eksperimen lanjutan dilakukan melalui penggerombolan representasi laten dan teknik node ablation untuk mengevaluasi kontribusi spasial tiap ruas jalan. Temuan mengungkapkan adanya ruas jalan dengan pengaruh signifikan terhadap prediksi, serta pola kemacetan khas berdasarkan gerombol geografis. Hasil penelitian ini memberikan kontribusi metodologis dan praktis dalam pemodelan kemacetan serta perencanaan transportasi berbasis data.
       
      This study aims to model and explore traffic congestion patterns in South Jakarta using the Spatio Temporal Graph Convolutional Network (ST-GCN) approach. The model integrates Graph Convolutional Networks to capture spatial relationships and Gated Recurrent Units to learn temporal dynamics. Traffic data were collected from Smart City Jakarta and combined with OpenStreetMap spatial data. After graph construction and preprocessing, the ST-GCN model was trained and evaluated against GRU and ST-GAT baseline models using MAE and SMAPE metrics. Results show that ST-GCN achieved reliable predictive performance with an average SMAPE of 8,70%. Further analysis involved latent representation clustering and node ablation experiments to evaluate the spatial contribution of each road segment. Findings reveal several critical roads with significant influence on prediction and distinct congestion patterns across spatial clusters. The outcomes offer both methodological and practical contributions to data-driven traffic modeling and urban transport planning.
       
      URI
      http://repository.ipb.ac.id/handle/123456789/169983
      Collections
      • UT - Statistics and Data Sciences [82]

      Copyright © 2020 Library of IPB University
      All rights reserved
      Contact Us | Send Feedback
      Indonesia DSpace Group 
      IPB University Scientific Repository
      UIN Syarif Hidayatullah Institutional Repository
      Universitas Jember Digital Repository
        

       

      Browse

      All of IPB RepositoryCollectionsBy Issue DateAuthorsTitlesSubjectsThis CollectionBy Issue DateAuthorsTitlesSubjects

      My Account

      Login

      Application

      google store

      Copyright © 2020 Library of IPB University
      All rights reserved
      Contact Us | Send Feedback
      Indonesia DSpace Group 
      IPB University Scientific Repository
      UIN Syarif Hidayatullah Institutional Repository
      Universitas Jember Digital Repository