Machine learning methods and agent-based models enable the optimization of the operation of high-capacity facilities. In this paper, we propose a method for automatically extracting and cleaning pedestrian traffic detector data for subsequent calibration of the ingress pedestrian model. The data was obtained from the waiting room traffic of a vaccination center. Walking speed distribution, the number of stops, the distribution of waiting times, and the locations of waiting points were extracted. Of the 9 machine learning algorithms, the random forest model achieved the highest accuracy in classifying valid data and noise. The proposed microscopic calibration allows for more accurate capacity assessment testing, procedural changes testing, and geometric modifications testing in parts of the facility adjacent to the calibrated parts. The results show that the proposed method achieves state-of-the-art performance on a violent-flows dataset. The proposed method has the potential to significantly improve the accuracy and efficiency of input model predictions and optimize the operation of high-capacity facilities.

Faculty of Civil Engineering Brno University of Technology Brno Czech Republic

Zobrazit více v PubMed

BAZJANAC V. Simulation of elevator performance in high-rise buildings under conditions of emergency. Conway DJ (ed) Human response to tall buildings Dowden. Stroudsburg, PA: Hutchinson & Ross, 1977, 316–328.

KULIGOWSKI Erica D., Richard D. PEACOCK a Bryan L. HOSKINS. A Review of Building Evacuation Models. Technical Note (NIST TN) - 1680. Gaithersburg, MD: National Institute of Standards and Technology, 2010.

LOVREGLIO Ruggiero, Enrico RONCHI a Michael J. KINSEY. An Online Survey of Pedestrian Evacuation Model Usage and Users. Fire Technology. Springer, 2020, 56, 1133–1153. Dostupné z: 10.1007/s10694-019-00923-8. DOI

BOYCE, Karen E., Steve M. GWYNNE V., ed. Engineering Data. In: HURLEY M. J. SFPE Handbook of Fire Protection Engineering. 5. Springer, 2016, s. 2429–2551. ISBN 978-1-4939-2565-0. Dostupné z: doi: 10.1007/978-1-4939-2565-0 DOI

THORNTON Charles, Richard O’KONSKI, B. HARDEMAN a Daniel SWENSON. Pathfinder: An Agent-Based Egress Simulator. 2011. Dostupné z: doi: 10.1007/978-1-4419-9725-8_94 DOI

BAHAMID, Alala a Azhar MOHD IBRAHIM. A review on crowd analysis of evacuation and abnormality detection based on machine learning systems. Neural Computing and Applications. 2022, (34), 21641–21655. Dostupné z: 10.1007/s00521-022-07758-5. DOI

JOHANSSON, Ander, Dirk HELBING a Pradyumn Kumar SHUKLA. Specification of the Social Force Pedestrian Model by Evolutionary Adjustment to Video Tracking Data. Advances in Complex Systems. 2007, 10(2), 271–288. Dostupné z: doi: 10.1142/S0219525907001355 DOI

YAO Zhenzhen, Guijuan ZHANG, Lu DIANJIE a Hong LIU. Data-driven crowd evacuation: a reinforcement learning method. Neurocomputing. Elsevier, 2019, (366), 314–327. ISSN 0925-2312. Dostupné z: doi: doi.org/10.1016/j.neucom.2019.08.021

WANG Q.H, LIU K. GAO a L. ZHANG. Improved Multi-Agent Reinforcement Learning for Path Planning-Based Crowd Simulation. IEEE Access. 2019, (7), 73841–73855. Dostupné z: doi: 10.1109/ACCESS.2019.2920913 DOI

LI Xinjin, Hong LIU, Junqing LI a Yan LI. Deep deterministic policy gradient algorithm for crowd-evacuation path planning. Computers & Industrial Engineering. 2021, (161). ISSN 0360-8352. Dostupné z: doi: 10.1016/j.cie.2021.107621 DOI

KIM, Beomjoon a Joelle PINEAU. Socially Adaptive Path Planning in Human Environments Using Inverse Reinforcement Learning. International Journal of Social Robotics. 2016, (8), 51–66. Dostupné z: 10.1007/s12369-015-0310-2. DOI

ALSALAT, Gehan Yahya, Mohammad EL-RAMLY a Aly Aly FAHMY. Detection of Mass Panic using Internet of Things and Machine Learning. International Journal of Advanced Computer Science and Applications. 2018, 9(5). Dostupné z: doi: 10.14569/IJACSA.2018.090542 DOI

YUAN Yuan, Jianwu FANG a Qi WANG. Online Anomaly Detection in Crowd Scenes via Structure Analysis. IEEE Transactions on Cybernetics. 2015, 45(3), 548–561. Dostupné z: doi: 10.1109/TCYB.2014.2330853 PubMed DOI

MARSDEN M., MCGUINNESS K., LITTLE a N S. O’CONNOR E. Holistic features for real-time crowd behaviour anomaly detection. IEEE International Conference on Image Processing (ICIP). 2016, 918–922. Dostupné z: doi: 10.1109/ICIP.2016.7532491 DOI

GUO Huiwen, Xinyu WU Nannan LI, Ruiqing FU, Guoyuan LIANG a Wei FENG. Anomaly detection and localization in crowded scenes using short-term trajectories. IEEE International Conference on Robotics and Biomimetics (ROBIO). 2013, 245–249. Dostupné z: doi: 10.1109/ROBIO.2013.6739466 DOI

CHAKER Rima, AL AGHBARI Zaher a JUNEJO Imran N. Social network model for crowd anomaly detection and localization. Pattern Recognition. 2017, 61, 266–281. ISSN 0031-3203. Dostupné z: doi: 10.1016/j.patcog.2016.06.016 DOI

GOLDHAMMER Michael, Konrad DOLL, Ulrich BRUNSMANN, André GENSLER a Bernhard SICK. Pedestrian’s Trajectory Forecast in Public Traffic with Artificial Neural Networks. 22nd International Conference on Pattern Recognition. 2014, 4110–4115. Dostupné z: doi: 10.1109/ICPR.2014.704 DOI

SCHULZ A. STIEFELHAGEN T a R. A Controlled Interactive Multiple Model Filter for Combined Pedestrian Intention Recognition and Path Prediction. IEEE 18th International Conference on Intelligent Transportation Systems. 2015, 173–178. Dostupné z: doi: 10.1109/ITSC.2015.37 DOI

ZHANG Yi, Zhe LI a Yongchao ZHANG. Validation and Calibration of an Agent-Based Model: A Surrogate Approach. Discrete Dynamics in Nature and Society. Hindawi, 2020. Dostupné z: 10.1155/2020/6946370. DOI

MAZIMPAKA, Jean Damascene a Sabine TIMPF. Trajectory data mining: A review of methods and applications. Journal of Spatial Information Science. 2016, 13, 61–99.

WANG Di, Tomio MIWA a Takayuki MORIKAWA. Big Trajectory Data Mining: A Survey of Methods, Applications, and Services. Sensors. 2020, 20(16). Dostupné z: 10.3390/s20164571. PubMed DOI PMC

KARIM Lamia, Azedine BOULMAKOUL a Karine ZEITOUNI. From Raw Pedestrian Trajectories to Semantic Graph Structured Model—Towards an end-to-end spatiotemporal analytics framework. Procedia Computer Science. 2021, 7(184), 60–67. Dostupné z: doi: 10.1016/j.procs.2021.03.018 DOI

WEI Ling-Yin, Yu ZHENG a Wen-Chih PENG. Constructing popular routes from uncertain trajectories. Proceedings of the 18th ACM SIGKDD international conference on Knowledge discovery and data mining. 2012, 195–203. Dostupné z: doi: 10.1145/2339530.2339562 DOI

DRIRA, Slah a Ian SMITH. A framework for occupancy detection and tracking using floor-vibration signals. Mechanical Systems and Signal Processing. 2022, 168. ISSN 0888-3270. Dostupné z: doi: 10.1016/j.ymssp.2021.108472 DOI

DRIRA S., REULAND Y., PAI S., NOH a I H. SMITH. Model-Based Occupant Tracking Using Slab-Vibration Measurements. Frontiers in Built Environment. 2019. Dostupné z: doi: 10.3389/fbuil.2019.00063 DOI

TOOHEY, Kevin a Matt DUCKHAM. Trajectory similarity measures. SIGSPATIAL Special. 7(1), 43–50. Dostupné z: 10.1145/2782759.2782767. DOI

YUAN Guan, Penghui SUN, Jie ZHAO, Daxing LI a Canwei WANG. A review of moving object trajectory clustering algorithms. Artificial Intelligence Review. 2016, 47, 123–144. Dostupné z: doi: 10.1007/s10462-016-9477-7 DOI

CAI Xumin, Berkay AYDIN, Anli JI a Rafal ANGRYK. A Framework for Local Outlier Detection from Spatio-Temporal Trajectory Datasets. 2020 25th International Conference on Pattern Recognition (ICPR). IEEE, 2021. ISSN 1051-4651. Dostupné z: doi: 10.1109/ICPR48806.2021.9412274 DOI

SAINI Rajkumar, artha Pratim ROY a Debi Prosad DOGRA. A segmental HMM based trajectory classification using genetic algorithm. Expert Systems with Applications. 2018, 93, 169–181. ISSN 0957-4174. Dostupné z: 10.1016/j.eswa.2017.10.021. DOI

DAVE, Rajesh N. Characterization and detection of noise in clusterin. Pattern Recognition Letters. 1991, 12(11), 657–664. ISSN 0167-8655. Dostupné z: 10.1016/0167-8655(91)90002-4. DOI

DUAN Yunna, Ying GAO Xiaojuan REN, Haoyuan CHE a Keyang YANG. Semi-supervised classification and noise detection. 2009 Sixth International Conference on Fuzzy Systems and Knowledge Discovery. IEEE, 2009. Dostupné z: doi: 10.1109/FSKD.2009.166 DOI

ESTER Martin, Hans-Peter KRIEGEL, Jörg SANDER a Xiaowei XU. A density-based algorithm for discovering clusters in large spatial databases with noise. Proceedings of the Second International Conference on Knowledge Discovery and Data Mining. 1996, 226–231.

BONACCORSO Giuseppe. Machine Learning Algorithms: A reference guide to popular algorithms for data science and machine learning. 1. Packt Publishing, 2017. ISBN 978–1785889622.

RAHMAN, Nadia a Imas Sukaesih SITANGGANG. Determination of Optimal Epsilon (Eps) Value on DBSCAN Algorithm to Clustering Data on Peatland Hotspots in Sumatra. IOP Conference Series: Earth and Environmental Science. 2016, (31). Dostupné z: doi: 10.1088/1755-1315/31/1/012012 DOI

MAATEN L. V. D. a Geoffrey E. HINTON. Visualizing Data using t-SNE. Journal of Machine Learning Research. 2008, 9, 2579–2605.

BRUCE Peter, Andrew BRUCE a Peter GEDEC. Practical Statistics for Data Scientists: 50+ Essential Concepts Using R and Python. 2nd edition. O’Reilly Media, 2020. ISBN 978–1492072942.

Haibo a GARCIA Edwardo A. Learning from Imbalanced Data. IEEE Transactions on Knowledge and Data Engineering. IEEE, 2009, 21(9), 1263–1284. ISSN 1558-2191. Dostupné z: doi: 10.1109/TKDE.2008.239 DOI

CHAWLA N. V., BOWYER K. W., HALL a W L. O. KEGELMEYER P. SMOTE: Synthetic Minority Over-sampling Technique. Journal Of Artificial Intelligence Research. 2002, 16, 321–357. Dostupné z: 10.48550/arXiv.1106.1813. DOI

GUAN Hongjiao, Yingtao ZHANG, Min XIAN H. D. CHENG a Xianglong TANG. SMOTE-WENN: Solving class imbalance and small sample problems by oversampling and distance scaling. Applied Intelligence. 2021, 51, 1394–1409. Dostupné z: 10.1007/s10489-020-01852-8. DOI

JIANG Z., T PAN C. ZHANG a J. YANG. A New Oversampling Method Based on the Classification Contribution Degree. Symmetry. 2021, 13. Dostupné z: 10.3390/sym13020194. DOI

ARAZO Eric, Diego ORTEGO Paul ALBERT, Noel E. O’CONNOR a Kevin MCGUINNESS. Pseudo-Labeling and Confirmation Bias in Deep Semi-Supervised Learning. Computer Vision and Pattern Recognition. 2020. Dostupné z: 10.48550/arXiv.1908.02983. DOI

RONCHI Enrico, Paul A. RENEKE a Richard PEACOCK. A Method for the Analysis of Behavioural Uncertainty in Evacuation Modelling. Fire Technology. 2014, (50), 1545–1571.