Updated on 2025/08/01

写真a

 
OKI Takuya
 
Organization
School of Environment and Society Associate Professor
Title
Associate Professor
External link

Degree

  • Dr. Eng. ( 2018.1   Tokyo Institute of Technology )

Research Areas

  • Social Infrastructure (Civil Engineering, Architecture, Disaster Prevention) / Architectural planning and city planning

Education

  • Tokyo Institute of Technology   Graduate School of Information Science and Engineering   Department of Mechanical and Environmental Informatics

    2011.4 - 2013.3

    　 More details

  • Tokyo Institute of Technology   School of Engineering   Dept. of Architecture and Building Engineering

    2007.4 - 2011.3

    　 More details

Research History

  • Tokyo Institute of Technology   School of Environment and Society   Assistant Professor

    2016.4 - 2018.12

    　 More details

  • Tokyo Institute of Technology   Graduate School of Science and Engineering, Department of Architecture and Building Engineering   Assistant Professor

    2013.4 - 2016.3

    　 More details

Professional Memberships

  • ARCHITECTURAL INSTITUTE OF JAPAN

    　 More details

  • GEOGRAPHIC INFORMATION SYSTEMS ASSOCIATION OF JAPAN

    　 More details

  • 災害情報学会

    　 More details

  • THE JAPANESE SOCIETY FOR ARTIFICIAL INTELLIGENCE

    　 More details

Papers

  • Webアンケートと保育所オープンデータベースを用いた世帯ベースの保育所選択モデルの構築

    増喜浩太郎, 沖拓弥

    GIS－理論と応用   Vol. 33 ( No. 1 )   2025.6

    　More details

    Language：Japanese   Publishing type：Research paper (scientific journal)  

    DOI： 10.5638/thagis.33.24

    researchmap

  • 実空間とVR空間における視覚的顕著性マップを用いた商店街歩行者の誘目性評価手法

    高井崚, 沖拓弥

    日本建築学会技術報告集   Vol. 31 ( No. 78 )   2025.6

    　More details

    Language：Japanese   Publishing type：Research paper (scientific journal)  

    DOI： 10.3130/aijt.31.1105

    researchmap

  • ワークプレイスはビッグデータの宝庫。そのデータ活用法とは

    沖拓弥

    NTTファシリティーズジャーナルDigital   2025.5

    　More details

    Language：Japanese  

    researchmap

  • Promoting the Integration of Public Perspective into Urban Design Decision- Making Based on Crowdsourced Visual Perception Method

    Lu Huang, Takuya Oki

    Landscape Architecture   Vol. 32 ( No. 5 )   2025.5

    　More details

    Publishing type：Research paper (scientific journal)  

    DOI： 10.3724/j.fjyl.LA20240101

    researchmap

  • Enhancing people’s walking preferences in street design through generative artificial intelligence and crowdsourcing surveys: The case of Tokyo

    Lu Huang, Takuya Oki

    Environment and Planning B: Urban Analytics and City Science   2025.3

    　More details

    Language：English   Publishing type：Research paper (scientific journal)  

    DOI： 10.1177/23998083251327405

    researchmap

  • Immersive gaze sharing for enhancing education: An exploration of user experience and future directions

    Yuval Kahlon, Weiheng Hu, Momoko Nakatani, Santosh Maurya, Takuya Oki, Jiang Zhu, Haruyuki Fujii

    Computers & Education: X Reality   2024.12

    　More details

    Publishing type：Research paper (scientific journal)  

    DOI： 10.1016/j.cexr.2024.100081

    researchmap

  • Evaluating the subjective perceptions of streetscapes using street-view images

    Yoshiki Ogawa, Takuya Oki, Chenbo Zhao, Yoshihide Sekimoto, Chihiro Shimizu

    Landscape and Urban Planning   Vol. 247   2024.7

    　More details

    Language：English   Publishing type：Research paper (scientific journal)  

    DOI： 10.1016/j.landurbplan.2024.105073

    researchmap

  • Unveiling the Non-Linear Influence of Eye-Level Streetscape Factors on Walking Preference: Evidence from Tokyo

    Lu Huang, Takuya Oki, Sachio Muto, Yoshiki Ogawa

    ISPRS International Journal of Geo-Information   Vol. 13 ( No. 4 )   2024.4

    　More details

    Language：English   Publishing type：Research paper (scientific journal)  

    DOI： 10.3390/ijgi13040131

    researchmap

  • データと人工知能（AI）、シミュレーションを活用し、少し異なる視点で防災・減災を考える

    沖拓弥

    JAEE NEWSLETTER   2023.12

    　More details

    Language：Japanese   Publisher：日本地震工学会  

    researchmap

  • Quantitative land price analysis via computer vision from street view images Reviewed

    Chenbo Zhao, Yoshiki Ogawa, Shenglong Chen, Takuya Oki, Yoshihide Sekimoto

    Engineering Applications of Artificial Intelligence   123   106294 - 106294   2023.8

    　More details

    Publishing type：Research paper (scientific journal)   Publisher：Elsevier BV  

    DOI： 10.1016/j.engappai.2023.106294

    researchmap

  • Automatic Evaluation of Street-Level Walkability Based on Computer Vision Techniques and Urban Big Data

    Lu Huang, Takuya Oki, Sachio Muto, Hongjik Kim, Yoshiki Ogawa, Yoshihide Sekimoto

    Intelligence for Future Cities   231 - 259   2023.6

    　More details

    Publishing type：Part of collection (book)   Publisher：Springer Nature Switzerland  

    DOI： 10.1007/978-3-031-31746-0_13

    researchmap

  • People Flow Trend Estimation Approach and Quantitative Explanation Based on the Scene Level Deep Learning of Street View Images

    Chenbo Zhao, Yoshiki Ogawa, Shenglong Chen, Takuya Oki, Yoshihide Sekimoto

    Remote Sensing   15 ( 5 )   1362 - 1362   2023.2

    　More details

    Publishing type：Research paper (scientific journal)   Publisher：MDPI AG  

    People flow trend estimation is crucial to traffic and urban safety planning and management. However, owing to privacy concerns, the collection of individual location data for people flow statistical analysis is difficult; thus, an alternative approach is urgently needed. Furthermore, the trend in people flow is reflected in streetscape factors, yet the relationship between them remains unclear in the existing literature. To address this, we propose an end-to-end deep-learning approach that combines street view images and human subjective score of each street view. For a more detailed people flow study, estimation and analysis were implemented using different time and movement patterns. Consequently, we achieved a 78% accuracy on the test set. We also implemented the gradient-weighted class activation mapping deep learning visualization and L1 based statistical methods and proposed a quantitative analysis approach to understand the land scape elements and subjective feeling of street view and to identify the effective elements for the people flow estimation based on a gradient impact method. In summary, this study provides a novel end-to-end people flow trend estimation approach and sheds light on the relationship between streetscape, human subjective feeling, and people flow trend, thereby making an important contribution to the evaluation of existing urban development.

    DOI： 10.3390/rs15051362

    researchmap

  • Deep Learning Approach for Classifying the Built Year and Structure of Individual Buildings by Automatically Linking Street View Images and GIS Building Data

    Yoshiki Ogawa, Chenbo Zhao, Takuya Oki, Shenglong Chen, Yoshihide Sekimoto

    IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing   2023.1

    　More details

    Language：English   Publishing type：Research paper (scientific journal)  

    DOI： 10.1109/JSTARS.2023.3237509

    researchmap

  • Predicting Impression Evaluation of Building Exterior Appearance Using Street Image Big Data and Deep Learning.

    Yusuke Imadegawa, Takuya Oki, Yoshiki Ogawa, Chenbo Zhao

    IEEE Big Data   1419 - 1428   2023

    　More details

    Publishing type：Research paper (international conference proceedings)  

    DOI： 10.1109/BigData59044.2023.10386316

    researchmap

    Other Link： https://dblp.uni-trier.de/db/conf/bigdataconf/bigdataconf2023.html#ImadegawaOOZ23

  • 保育所選択行動モデルの構築と地域保育施設計画評価への応用

    沖拓弥, 増喜浩太郎

    日本建築学会計画系論文集   Vol. 87 ( No. 802 )   2022.12

    　More details

    Language：Japanese   Publishing type：Research paper (scientific journal)  

    DOI： 10.3130/aija.87.2480

    researchmap

  • 画像ビッグデータと大規模被験者アンケートに基づく住宅地における街路印象評価推定モデル

    沖拓弥, 木澤佐椰茄

    日本建築学会計画系論文集   Vol. 87 ( No. 800 )   2022.10

    　More details

    Language：Japanese   Publishing type：Research paper (scientific journal)  

    DOI： 10.3130/aija.87.2102

    researchmap

  • ESTIMATION OF THE NUMBER OF WALKING PEOPLE IN STAIRCASES USING INFRARED HUMAN DETECTION SENSORS

    OSARAGI Toshihiro, YOSHIYUKI Natsumi, KISHIMOTO Maki, OKI Takuya, IYAMA Jun, FUKUSHIMA Yoshihiro

    Journal of Architecture and Planning (Transactions of AIJ)   87 ( 799 )   1699 - 1707   2022.9

    　More details

    Language：Japanese   Publishing type：Research paper (scientific journal)   Publisher：Architectural Institute of Japan  

    The spatial distribution of people in a building could provide valuable information for an effective guidance in emergency. This paper aims to develop a method to estimate the number of walking people in staircases using infrared human detection sensors. First, we constructed a walking behavior model and a sensor response model using the survey data observed in an emergency drill. Next, using these two models, we generated a large volume training data on human flows and corresponding sensor responses for a neural network model. Finally, we demonstrated the validation of the model through the comparison with the survey data.

    DOI： 10.3130/aija.87.1699

    CiNii Research

    researchmap

  • A METHOD FOR REGIONAL ANALYSIS USING DEEP LEARNING BASED ON BIG DATA OF OMNIDIRECTIONAL IMAGES OF STREETS

    T. Oki, Y. Ogawa

    International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences - ISPRS Archives   43 ( B4-2022 )   545 - 552   2022.5

    　More details

    Publishing type：Research paper (international conference proceedings)  

    DOI： 10.5194/isprs-archives-XLIII-B4-2022-545-2022

    Scopus

    researchmap

  • Research conducted in the Dr. Takuya Oki Laboratory

    Takuya Oki

    Impact   2022.2

    　More details

    Language：English   Publisher：Science Impact Ltd  

    DOI： 10.21820/23987073.2022.1.18

    researchmap

  • Joining street-view images and building footprint gis data

    Yoshiki Ogawa, Takuya Oki, Shenglong Chen, Yoshihide Sekimoto

    Proceedings of the 1st ACM SIGSPATIAL International Workshop on Searching and Mining Large Collections of Geospatial Data, GeoSearch 2021   18 - 24   2021.11

    　More details

    Publishing type：Research paper (international conference proceedings)  

    DOI： 10.1145/3486640.3491395

    Scopus

    researchmap

  • 空間五感 世界の建築・都市デザイン

    沖拓弥, and 共著

    2021.3

    　More details

    Language：Japanese   Publisher：井上書院  

    researchmap

  • Development of a Data-driven Evacuee Distribution Prediction Model by Integrating Graph Neural Networks and Data Assimilation

    TAKASAKI Dan, OGAWA Yoshiki, IMAIZUMI Masaaki, OKI Takuya, OYAMA Yuki

    Proceedings of the Annual Conference of JSAI   JSAI2021   2D1OS602 - 2D1OS602   2021

    　More details

    Language：Japanese   Publisher：The Japanese Society for Artificial Intelligence  

    In the disaster response immediately after a large earthquake, it is important to grasp the real-time situation of people's stay and evacuation. However, it is a challenge to shorten the computation time and improve the prediction accuracy. This paper aims to build a new data-driven evacuee distribution prediction model, which integrates graph neural network (GNN) and data assimilation, to efficiently and accurately predict the distribution of the number of evacuees per road link and time unit: (1)We train a GNN model to obtain a likelihood distribution corresponding to the probability of movement from a node to its connecting links; (2) We apply a particle filter, one of the data assimilation methods, to the probabilistic predictive evacuee distribution obtained from the likelihood distribution mentioned above. The sequential calculation by the model is expected to improve the speed and accuracy of evacuee distribution prediction.

    DOI： 10.11517/pjsai.jsai2021.0_2d1os602

    CiNii Research

    researchmap

  • Model for Estimation of Building Structure and Built Year Using Building Facade Images and Attributes Obtained from a Real Estate Database

    Takuya Oki, Yoshiki Ogawa

    Urban Book Series   549 - 573   2021

    　More details

    Publishing type：Part of collection (book)  

    DOI： 10.1007/978-3-030-76059-5_27

    Scopus

    researchmap

  • Exploring the heterogeneity of human urban movements using geo-tagged tweets Reviewed

    Ding Ma, Toshihiro Osaragi, Takuya Oki, Bin Jiang

    International Journal of Geographical Information Science   Vol. 34 ( issue 12 )   2020.1

    　More details

    Language：English   Publishing type：Research paper (scientific journal)  

    DOI： 10.1080/13658816.2020.1718153

    researchmap

  • 想定東京湾北部地震時における避難所の混雑度について Reviewed

    大佛 俊泰, 荻野 光司, 廣川 典昭, 沖 拓弥

    日本建築学会計画系論文集 計画系   Vol. 84 ( No. 760 )   2019.6

    　More details

    Language：Japanese   Publishing type：Research paper (scientific journal)  

    DOI： 10.3130/aija.84.1521

    researchmap

  • Quick Estimation Method of Property Damage and Human Casualty in the Event of a Large Earthquake Reviewed

    Takuya Oki, Toshihiro Osaragi, Yoho Sakamoto

    Journal of Disaster Research (Special Issue on Disaster and Big Data 4)   Vol. 14 ( No. 3 )   2019.3

    　More details

    Language：English   Publishing type：Research paper (scientific journal)  

    DOI： 10.20965/jdr.2019.p0478

    researchmap

  • Disaster Response and Mitigation Support Technology for All-Hazards in Tokyo Metropolitan Area Reviewed

    Yoshiaki Hisada, Toshihiro Osaragi, Masahiro Murakami, Osamu Mizuno, Wataru Kobayashi, Susumu Yasuda, Miho Ohara, Tomohisa Yamashita, Kazuyuki Takada, Takashi Suematsu, Jun Shindo, Takuya Oki, Akira Kakizaki

    Journal of Disaster Research   Vol. 14 ( No. 2 )   2019.3

    　More details

    Language：English   Publishing type：Research paper (scientific journal)   Publisher：Journal of Disaster Research (Special Issue on Disaster and Big Data 3)  

    researchmap

  • 災害時における救援救護活動のための効率的な地域内巡回について Reviewed

    大佛俊泰, 木村 将, 沖 拓弥

    日本建築学会計画系論文集 計画系   Vol. 84 ( No. 757 )   2019.3

    　More details

    Language：Japanese   Publishing type：Research paper (scientific journal)  

    DOI： 10.3130/aija.84.743

    researchmap

  • Efficient regional travel for rescue and relief activities in a disaster Reviewed

    Toshihiro Osaragi, Masashi Kimura, Takuya Oki

    Lecture Notes in Geoinformation and Cartography   425 - 441   2019

    　More details

    Publishing type：Research paper (international conference proceedings)  

    DOI： 10.1007/978-3-030-19424-6_23

    Scopus

    researchmap

  • SNS上のテキスト情報に基づく群集混雑度の定量化手法の検討:東日本大震災時のツイートを用いたアンケート調査結果の分析 Reviewed

    沖拓弥

    災害情報   ( No. 16-2 )   283‐294   2018.7

    　More details

    Language：Japanese   Publishing type：Research paper (scientific journal)  

    J-GLOBAL

    researchmap

  • Assessment of Street Network Accessibility in Tokyo Metropolitan Area after a Large Earthquake Reviewed

    Toshihiro Osaragi, Maki Kishimoto, Takuya Oki

    Journal of Disaster Research (Special Issue on Disaster and Big Data 3)   Vol. 13 ( No. 2 )   262 - 271   2018.3

    　More details

    Language：English   Publishing type：Research paper (scientific journal)  

    © 2018, Fuji Technology Press. All rights reserved. It is difficult to evaluate the street network accessibility after a large earthquake occurs. In this paper, we construct a model to evaluate the street network accessibility for wide-area emergency behaviors under the condition of property damage in the Tokyo Metropolitan Area after a large earthquake. Additionally, we analyze the relationships between a local environment and street network accessibility by using multiple regression analysis. Finally, we discuss some important factors for evaluating risk mitigation strategies.

    DOI： 10.20965/jdr.2018.p0262

    researchmap

  • Effects of Firefighting Activities Performed by Local Residents to Mitigate Fire Destruction Damage and Human Casualties in Large Earthquakes Reviewed

    Takuya Oki, Toshihiro Osaragi

    Journal of Disaster Research (Special Issue on Disaster and Big Data 3)   Vol. 13 ( No. 2 )   272 - 280   2018.3

    　More details

    Language：English   Publishing type：Research paper (scientific journal)  

    © 2018, Fuji Technology Press. All rights reserved. Fire-spread prevention activities, which are performed by local residents in an early phase of fire, play an important role in reducing the destruction damage caused by fire in large earthquakes. However, few studies have focused on the fire-spread prevention activities that are carried out in the confused situation immediately following a large earthquake. Therefore, to date, there has been no sufficiently comprehensive discussion regarding the effectiveness of such activities. In this study, we estimate the possibility that local residents can prevent the spread of fire from the building of fire origin to an adjacent building by using equipment such as stand pipes. For this purpose, we utilize the agent-based simulator of property damage and human behavior at the time of a large earthquake that we developed in an earlier study. In addition, we demonstrate the effects of some measures for increasing the success rate of preventing fire-spread by comparing the simulation results under the following assumptions: (1) The percentage of streets blocked by the rubble of collapsed buildings is decreased, (2) the number of stand pipes is

    DOI： 10.20965/jdr.2018.p0272

    researchmap

  • 大地震時の物的被害を考慮した人間行動シミュレーションによる木造住宅密集地域の減災対策評価に関する研究 Reviewed

    沖拓弥

    2018.1

    　More details

    Language：Japanese   Publishing type：Doctoral thesis  

    researchmap

  • A REAL TIME SYNCHRONOUS SYSTEM FOR COLLECTING, SHARING, AND UTILIZING DISASTER INFORMATION Reviewed

    大佛俊泰, 沖拓弥

    日本建築学会計画系論文集   Vol. 82 ( No. 739 )   2451‐2459(J‐STAGE) - 2459   2017.9

    　More details

    Language：Japanese   Publishing type：Research paper (scientific journal)   Publisher：日本建築学会  

     When a large earthquake occurs, rescue operations and fire-fighting are obstructed by street-blockages. In order to reduce property/human damages caused by the delay in arrival at the disaster site, it is important to quickly collect, share and utilize disaster information among multiple users. In this paper, we developed a system that can collect and share information acquired by users in real time in the event of a disaster. The advantages of the system are summarized as follows: (1) Users can access to the system by using various kinds of information terminals such as mobile phones (Android/iOS) and personal computers because it was implemented as a Web application; (2) The system can run on a cloud server in several countries with high disaster resistance; (3) A simulation to predict the property damage by fire-spreading can be performed based on collected information on the location of building fires as an example of secondary usage of disaster information; (4) Persons in charge of collecting disaster information and the location with high possibility of damages are recommended to support users for effective information collecting.<br> Using this system, we conducte

    DOI： 10.3130/aija.82.2451

    J-GLOBAL

    researchmap

  • CHOICE BEHAVIOR OF STOPOVER-STATION BY RAILWAY USERS ON THEIR WAY HOME Reviewed

    大佛俊泰, 岩淵紗葵, 沖拓弥

    日本建築学会計画系論文集   Vol. 82 ( No. 736 )   1561‐1568(J‐STAGE) - 1568   2017.6

    　More details

    Language：Japanese   Publishing type：Research paper (scientific journal)   Publisher：日本建築学会  

     Railway commuters to work place/school, who are the majority of railway users, often make stopover at various stations for eating or shopping on their way home. In recent years, the mobility of railway users has been diversified by expansion of railway network and redevelopment around railway stations. Therefore, we need to analyze such stopover behavior by commuters in order not only to improve safe and comfortableness of railway stations but also to activate the commercial activities in the vicinity of stations.<br> In this paper, we constructed a multinomial logit model, which describes commuters' choice behavior of stopover-stations for eating or shopping on their way home from their workplace. More specifically, we considered commercial facilities in the vicinity of the stopover-station and the positional relationships between their home and workplace as explanatory variables. Analyzing the person trip data obtained in three different time (1988, 1998, and 2008) in Tokyo Metropolitan Area by the proposed model, we demonstrated the estimation accuracy of the model and the attractiveness of stopover-stations quantitatively. On the basis of the estimated parameters, we

    DOI： 10.3130/aija.82.1561

    J-GLOBAL

    researchmap

  • Wide-Area Evacuation Simulation Incorporating Rescue and Firefighting by Local Residents Reviewed

    Toshihiro Osaragi, Takuya Oki

    Journal of Disaster Research (Special Issue on Disaster and Big Data)   Vol. 12 ( No. 2 )   2017.3

    　More details

    Language：English   Publishing type：Research paper (scientific journal)  

    DOI： 10.20965/jdr.2017.p0296

    researchmap

  • Parameter Smoothing Specification in Bayesian Geographically and Temporally Weighted Regression: Case Study of Analyzing Spatio-temporal Distribution of Population in Tokyo Metropolitan Area Reviewed

    Sajad Alipoureshliki, Toshihiro Osaragi, TakuyaOki

    Theory and Applications of GIS   Vol. 24 ( No. 2 )   73 - 84   2017.1

    　More details

    Language：English   Publishing type：Research paper (scientific journal)   Publisher：地理情報システム学会 ; [1993]-  

    researchmap

  • Evaluation of Conversion to Quake-Resistant Buildings in Terms of Wide-Area Evacuation and Fire-Brigade Accessibility. Reviewed

    Takuya Oki, Toshihiro Osaragi

    2017

    　More details

    Publishing type：Research paper (international conference proceedings)  

    researchmap

  • Urban improvement policies for reducing human damage in a large earthquake by using wide-area evacuation simulation incorporating rescue and firefighting by local residents Reviewed

    Takuya Oki, Toshihiro Osaragi

    Lecture Notes in Geoinformation and Cartography   4   449 - 468   2017

    　More details

    Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：Springer Berlin Heidelberg  

    DOI： 10.1007/978-3-319-57819-4_25

    Scopus

    researchmap

  • SIMULATION ANALYSIS OF WIDE-AREA EVACUATION IN DENSELY BUILT-UP WOODEN RESIDENTIAL AREAS AFTER A LARGE EARTHQUAKE INCORPORATING RESCUE ACTIVITIES BY LOCAL RESIDENTS Reviewed

    沖拓弥, 大佛俊泰

    日本建築学会計画系論文集   Vol. 81 ( No. 724 )   1345‐1353 - 1353   2016.6

    　More details

    Language：Japanese   Publishing type：Research paper (scientific journal)   Publisher：日本建築学会  

     In order to reduce human casualties, rescue activities by local residents play an important role in the chaotic times immediately after an earthquake occurs because it is difficult to expect the sufficient and quick support by public rescue teams such as firefighters, police officers and the Self-Defense Forces. Therefore, it is necessary to quantitatively understand both the effects and risk of these activities based on the numerical analyses, and concretely discuss how to perform rescue activities by local residents as part of disaster mitigation planning.<br> In this paper, we incorporate the following aspects into the wide-area evacuation simulation model which we have previously developed: (1) people trapped in a building and on a street; (2) rescue activities by local residents; (3) the behavior of people who get over the rubble of collapsed buildings; (4) the difference of the attribute of evacuees. Using this simulation model, first, we demonstrate the effectiveness of rescue activities by local residents in densely built-up wooden residential areas in the aftermath of a large earthquake. Even considering the influence of property damages (such as building-collap

    DOI： 10.3130/aija.81.1345

    J-GLOBAL

    researchmap

  • Modeling the Spatio-Temporal Relationships for The Distribution of Population with Geographically and Temporally Weighted Regression Reviewed

    Sajad, Toshihiro Osaragi, Takuya Oki

    Journal of Architecture and Planning (Transactions of AIJ)   Vol. 81 ( No. 723 )   1165 - 1173   2016.5

    　More details

    Language：English   Publishing type：Research paper (scientific journal)   Publisher：日本建築学会  

     This paper applies a spatio-temporal model to assess the variability of local parameters associated with the urban population change over time. With the Person Trip survey dataset and the detailed digital map of land use, Geographically and Temporally Weighted Regression (GTWR) is used to analyze the relationship between distributions of occupants inside residential buildings and commercial buildings in Tokyo Metropolitan Area (TMA). The outcome of the models has shown that GTWR has higher explanatory power (R<sup>2</sup>=95%) compared to global regression model (R<sup>2</sup>=0.5%) and Geographically Weighted Regression (R<sup>2</sup>=91%). Furthermore, there is a significant pattern of spatio-temporal characteristics for distribution of populations at the inner city, its neighborhoods and suburbs of Tokyo.

    DOI： 10.3130/aija.81.1165

    researchmap

  • Modeling Human Behavior of Local Residents in the Aftermath of a Large Earthquake – Wide-Area Evacuation, Rescue and Firefighting in Densely Built-Up Wooden Residential Areas Reviewed

    Takuya Oki, Toshihiro Osaragi

    Journal of Disaster Research (Special Issue on Disaster and Big Data)   Vol. 11 ( No. 2 )   2016.3

    　More details

    Language：English   Publishing type：Research paper (scientific journal)  

    DOI： 10.20965/jdr.2016.p0188

    researchmap

  • Wide-area Evacuation Difficulty in Densely-built Wooden Residential Areas. Reviewed

    Takuya Oki, Toshihiro Osaragi

    13th Proceedings of the International Conference on Information Systems for Crisis Response and Management, Rio de Janeiro, Brasil, May 22-25, 2016.   2016

    　More details

    Publishing type：Research paper (international conference proceedings)   Publisher：ISCRAM Association  

    researchmap

  • INFORMATION COLLECTION OF STREET BLOCKAGE AFTER A LARGE EARTHQUAKE FOR REDUCING ACCESS TIME OF FIRE FIGHTERS Reviewed

    大佛俊泰, 廣川典昭, 沖拓弥

    日本建築学会計画系論文集   Vol. 80 ( No. 709 )   465 - 473   2015.3

    　More details

    Language：Japanese   Publishing type：Research paper (scientific journal)   Publisher：日本建築学会  

    The previous studies presume that the use of emergency vehicles will be obstructed by collapsed buildings after a large earthquake. In this paper, we investigate a method for assisting fire fighters for their effective and efficient access to the locations of fires. More specifically, first, we construct a simulation model that describes the movement of fire engines under the condition of street-blockages caused by collapsed buildings. Next, using the proposed model, we attempt to evaluate the effects of street-blockage information collected by residents on improving the accessibility of fire engines. Finally, based on the simulation result, we demonstrate that the access time to the locations of fires can be effectively reduced by using the street-blockage information.

    DOI： 10.3130/aija.80.465

    J-GLOBAL

    researchmap

  • Development of Real Time Synchronous Web Application for Posting and Utilizing Disaster Information. Reviewed

    Ikki Niwa, Toshihiro Osaragi, Takuya Oki, Noriaki Hirokawa

    12th Proceedings of the International Conference on Information Systems for Crisis Response and Management, Krystiansand, Norway, May 24-27, 2015.   2015

    　More details

    Publishing type：Research paper (international conference proceedings)   Publisher：ISCRAM Association  

    researchmap

  • Influence of Information-Hearsay on Wide-Area Evacuation at a Large Earthquake. Reviewed

    Takuya Tsuchiya, Toshihiro Osaragi, Takuya Oki

    12th Proceedings of the International Conference on Information Systems for Crisis Response and Management, Krystiansand, Norway, May 24-27, 2015.   2015

    　More details

    Publishing type：Research paper (international conference proceedings)   Publisher：ISCRAM Association  

    researchmap

  • EVALUATION OF IMPROVEMENT PROJECTS IN DENSELY BUILT-UP WOODEN RESIDENTIAL AREA IN TERMS OF DIFFICULTY IN WIDE-AREA EVACUATION Reviewed

    大佛俊泰, 沖拓弥

    日本建築学会計画系論文集   Vol. 79 ( No. 696 )   437 - 444   2014.2

    　More details

    Language：Japanese   Publishing type：Research paper (scientific journal)   Publisher：日本建築学会  

    DOI： 10.3130/aija.79.437

    J-GLOBAL

    researchmap

  • EVALUATION OF IMPROVEMENT PROJECTS IN DENSELY BUILT-UP WOODEN RESIDENTIAL AREA IN TERMS OF DIFFICULTY IN WIDE-AREA EVACUATION

    OSARAGI Toshihiro, OKI Takuya

    Journal of Architecture and Planning (Transactions of AIJ)   79 ( 696 )   437 - 444   2014

    　More details

    Language：Japanese   Publisher：日本建築学会  

    In high-density wooden residential areas, some aspects of spatial characteristics must be improved to safeguard against high risks which may occur in a major earthquake. However, improvements have not been steadily performed due to the lack of cost-benefit analysis on improvement projects. In this paper, we evaluated past improvement projects in terms of difficulty in wide-area evacuation by using a multi-agent simulation model. Results showed that some buildings where residents have difficulty in evacuating have not been improved by these projects, and this risk can be reduced by shortening the street length which these buildings are adjacent to.

    CiNii Books

    researchmap

  • DIFFICULTY OF EVACUATION IN DENSELY BUILT-UP AREA IN THE AFTERMATH OF A LARGE-SCALE EARTHQUAKE Reviewed

    大佛俊泰, 沖拓弥

    日本建築学会計画系論文集   Vol. 77 ( No. 681 )   2561 - 2567   2012.11

    　More details

    Language：Japanese   Publishing type：Research paper (scientific journal)   Publisher：日本建築学会  

    DOI： 10.3130/aija.77.2561

    J-GLOBAL

    researchmap

  • DIFFICULTY OF EVACUATION IN DENSELY BUILT-UP AREA IN THE AFTERMATH OF A LARGE-SCALE EARTHQUAKE

    OSARAGI Toshihiro, OKI Takuya

    Journal of Architecture and Planning (Transactions of AIJ)   77 ( 681 )   2561 - 2567   2012

    　More details

    Language：Japanese   Publisher：日本建築学会  

    In the field of disaster prevention planning, it is highly important to consider the difficulty of evacuation. In order to evaluate difficulty of evacuation dealing with any possible damages by a large-scale earthquake, we construct a simulation model, which describes physical damages (such as house-collapse, city-fire, and road-blockage) and people's behavior in terms of difficulty in wide-area evacuation. Based on the results of agent based simulations, we demonstrated some new findings: (1) the difficulty in wide-area evacuation has a significant correlation respectively with the density of wooden houses and the rate of road-blockage, (2) some areas might have difficulties even if they are located near the evacuation site, (3) the number of people with difficulty varies according to occurrence-time of disaster, the characteristic of areas, and the unfavorable situation for evacuation.

    CiNii Books

    researchmap

▼display all

MISC

▼display all

Presentations

▼display all

Awards

  • ISCRAM2017 Best Research Paper Award

    2017.5   ISCRAM  

    Takuya Oki, Toshihiro Osaragi

    　More details

  • 関東支部研究発表会 優秀発表

    2015.3   日本建築学会  

    沖 拓弥

    　More details

  • 都市計画部門 若手優秀発表

    2014.9   日本建築学会  

    沖 拓弥

    　More details

  • 建築学科同窓会 冬夏賞（修士）

    2014.1   東京工業大学  

    沖 拓弥

    　More details

  • 情報環境学専攻賞

    2013.3   東京工業大学  

    沖 拓弥

    　More details

  • 学生論文コンテスト 修士論文部門最優秀賞

    2013.3   都市住宅学会  

    沖 拓弥

    　More details

▼display all

Research Projects

  • Optimization of hospital ICUs achieved through flow line analysis and flow prediction using small LiDAR sensors and cutting-edge AI technology

    Grant number：25K01391  2025.4 - 2030.3

    Japan Society for the Promotion of Science  Grants-in-Aid for Scientific Research  Grant-in-Aid for Scientific Research (B)

    　 More details

    Grant amount：\25480000 （ Direct Cost: \19600000 、 Indirect Cost：\5880000 ）

    researchmap

  • Development of Housing Environment Assessment Methods Using Diverse Big Data and Artificial Intelligence for the Formation of Sustainable Residential Areas

    Grant number：22K04490  2022.4 - 2025.3

    Japan Society for the Promotion of Science  Grants-in-Aid for Scientific Research  Grant-in-Aid for Scientific Research (C)

    　 More details

    Grant amount：\4030000 （ Direct Cost: \3100000 、 Indirect Cost：\930000 ）

    researchmap

  • 「観測ビッグデータ同化」による大地震時のリアルタイム避難者分布推定技術の開発

    Grant number：19K04946  2019.4 - 2023.3

    日本学術振興会  科学研究費助成事業  基盤研究(C)

    沖 拓弥, 小川 芳樹, 今泉 允聡, 大山 雄己, 大佛 俊泰

    　 More details

    Grant amount：\4420000 （ Direct Cost: \3400000 、 Indirect Cost：\1020000 ）

    2020年度には，Gretel [Cordonnier et al. 2019]と呼ばれるグラフニューラルネットワーク（GNN）モデルと，データ同化手法の一つであるパーティクルフィルタを統合した，データ駆動型の新たな避難者分布予測モデルを提案した。
    2021年度は，上述の避難者分布予測モデルを改良するとともに，擬似的な観測避難者分布の生成に用いていているエージェントベースシミュレーションの結果と比較することで，主に計算時間と予測精度の観点からモデルの有用性を検証することを試みた。
    具体的には，第一に，2020年度時点では十分に考慮できていなかった時刻情報を考慮できるよう，入力データの仕様や計算プログラムの改良を行った。時刻情報は，データ同化を行うにあたって重要であるため，発生からの経過時間に応じた道路閉塞状況のみならず，各避難者の位置情報も時刻情報とセットで扱うようにした。加えて，ベースとして用いているGretelは局所的な状況に基づき将来予測を行うものであるが，この特性は目的地（避難場所）をもって行動する避難行動とは乖離がある。そこで，道路ネットワーク上の各交差点（ノード）から各避難場所までの距離をノードの特徴量として組み込むことで，より大局的な避難行動の傾向を考慮した予測を可能とすることを目指した。
    また，GNNに学習させる避難行動軌跡数やデータ同化で用いるパーティクル数と，計算コストや予測精度との関係を分析し，リアルタイム予測に求められる要件や期待できる精度について検討を試みた。

    researchmap

  • Prediction method of spatiotemporal distribution of transient occupants and risk reduction technique after a disaster

    Grant number：17H00843  2017.4 - 2021.3

    Japan Society for the Promotion of Science  Grants-in-Aid for Scientific Research  Grant-in-Aid for Scientific Research (A)

    Osaragi Toshihiro

    　 More details

    Grant amount：\31980000 （ Direct Cost: \24600000 、 Indirect Cost：\7380000 ）

    In order to reduce the confusion and secondary damage such as crowd accidents after a disaster in city areas, first of all, it is essential to obtain the precise spatiotemporal demographic data that includes detailed information such as personal attributes (age, gender, occupation), and information on position (place, building use) and time (season, day of the week, time). In this research project, we constructed some novel methods to generate the detailed spatiotemporal databases of transient occupants in city areas by combining multiple vital existing demographic data. Also we have developed simulation models to predict the movement of population after a large earthquake. Based on these methods and models, we have developed a technology to support the control of damage and confusion after a disaster.

    researchmap

  • Relationship between linguistic expression and human behavior for estimating crowd congestion degree after a large earthquake using Twitter data

    Grant number：15K18176  2015.4 - 2018.3

    Japan Society for the Promotion of Science  Grants-in-Aid for Scientific Research  Grant-in-Aid for Young Scientists (B)

    Oki Takuya

    　 More details

    Grant amount：\4160000 （ Direct Cost: \3200000 、 Indirect Cost：\960000 ）

    Using the tweets and photographs posted by users confronted with the crowd congestion, the possibility of grasping the degree of congestion in the facilities (such as railway stations), on the railway lines, or on the roads in real time was discussed.
    More specifically, using tweets written in Japanese posted before and after the Great East Japan Earthquake (for 24 hours since 2:00 p.m. on March 11, 2011) as a case study, the method to extract tweets related to crowded conditions and the usefulness of tweets for grasping the time transition and spatial distribution of crowd congestion were demonstrated. Furthermore, conducting a Web questionnaire survey, the relationships between the actual congestion degree and the linguistic expressions about the congestion level and the selected behaviors were clarified.

    researchmap

  • Study on Improvement Projects in Densely Built-up Wooden Residential Area from the Viewpoint of Difficulty in Wide-area Evacuation at a Large Earthquake

    Grant number：25889023  2013.8 - 2015.3

    Japan Society for the Promotion of Science  Grants-in-Aid for Scientific Research  Grant-in-Aid for Research Activity Start-up

    OKI Takuya

    　 More details

    Grant amount：\1300000 （ Direct Cost: \1000000 、 Indirect Cost：\300000 ）

    The various analyses related to the evaluation of improvement projects in densely built-up wooden residential area were performed using the simulation model, which describes wide-area evacuation at a large earthquake. Firstly, the past improvement projects were evaluated with the difficulty in wide-area evacuation between two time points (1991 and 2006) as an index. Secondly, as an example of the application of the simulation model to considering the specific methods to improve cities, the effects of adding new evacuation routes connecting between two narrow and long streets were verified. Thirdly, the exposure risk with fire-spreading for people and the congestion of streets and evacuation areas were estimated considering people who return home on foot after an earthquake occurs. Fourthly, the effects and risk of rescue activities by local residents were quantified. Lastly, the basic analyses were performed in order to take automobile users at a large earthquake into account.

    researchmap

▼display all