APPLICATION OF NEURAL NETWORK MODELING TO FORECASTING THE VEHICLE STAY DURATION AT THE PUBLIC TRANSPORT STOP
Abstract
Introduction. Increasing the demand for public transport among the urban population can be achieved through a set of various measures, one of which is to improve the transportation system and improve the quality of passenger service at various stages of the transportation process. Modern methods of elaboration and analysis of parameters of functioning of transport systems allow to estimate the influence of various factors on transport processes and to predict results of such influence. Most transport processes have a stochastic, nonlinear structure. In such cases, it is advisable to use methods of artificial intelligence, in particular artificial neural networks. Purpose. The purpose of the article is to determine the duration of vehicle stay at a public transport stop, using neural network modeling. Results. The basic principles of functioning of artificial neural networks and rules of their use are revealed in the work. The expediency of using neural network modeling to predict the stay duration of a vehicle at public transport stops is analyzed. In particular, the influence of the following factors was analyzed: route length, distance from the beginning of the route to the researched stop, the interval between vehicles of a certain route and passenger exchange at the stop. Based on the information collected during field observations, a neural network was created and the duration of the vehicle’s stay at the stop was predicted in the Deductor software environment. The quality of the obtained model was evaluated. Conclusions. Neural network modeling is an effective tool for studying transport processes. The obtained results testify to the sufficient accuracy of the obtained model (the average stay duration of the vehicle at the stop is 24 s in the morning and 21 s in the lunch period, deviation in the range from 5 to 9.6 %). Further research will focus on improving the accuracy of the model by, in particular, expanding the list of input parameters.
Downloads
References
2. Кириченко А. А. Нейропакеты – современный интеллектуальный инструмент исследователя. 2013. 297 с.
3. Waleed Khalid Shihab, Sivaram Prasad. Artificial neural network. International Journal of Scientific Research. 2022. Vol. 8, Issue 7, pp. 18307–18328. DOI: 10.24327/ijrsr.2017.0807.0485
4. Вакуленко К. Є., Доля К. В. Управління міським пасажирським транспортом : навчальний посібник. Харків. нац. ун-т міськ. госп-ва ім. О. М. Бекетова. Харків, 2015. 257 с.
5. Півторак Г. В. Дослідження тривалості простою маршрутних транспортних засобів на зупинках громадського транспорту Львова. Автошляховик України. 2019. № 2. С. 17–23.
6. Яблоков І. В. Логістичні принципи прогнозування розподілу транспортних потоків на основі нейронних мереж. Економіко-математичне моделювання соціально-економічних систем. 2009. Вип. 14. С. 144–149.
7. Kang D., Lv Y. and Chen Y. Short-term traffic flow prediction with LSTM recurrent neural network. IEEE 20th International Conference on Intelligent Transportation Systems (ITSC). 2017. Pp. 1–6. DOI: 10.1109/ITSC.2017.8317872
8. Panovski D., Scurtu V. and Zaharia T. A Neural Network-based Approach for Public Transportation Prediction with Traffic Density Matrix. 7th European Workshop on Visual Information Processing (EUVIP). 2018. Pp. 1–6. DOI: 10.1109/EUVIP.2018.8611683
9. Junchen Ye, Leilei Sun, Bowen Du, Yanjie Fu, Xinran Tong, Hui Xiong. Co-Prediction of Multiple Transportation Demands Based on Deep Spatio-Temporal Neural Network. In Proceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining (KDD‘19). 2019. Association for Computing Machinery, New York, USA, 305–313. DOI: https://doi.org/10.1145/3292500.3330887
10. Liu Guojin, Yin Zhenzhi, JiaYunjian, Xie Yulaib. Passenger flow estimation based on convolutional neural network in public transportation system. Knowledge-Based Systems. 2017. Volume 123, 1. P. 102–115.
11. Shuwei Wang, Ronggui Zhou, Lin Zhao. Forecasting Beijing Transportation Hub Areas’s Pedestrian Flow Using Modular Neural Network. Discrete Dynamics in Nature and Society. 2015. Vol. 2015, Article ID 749181. 6 p. URL: https://doi.org/10.1155/2015/749181
12. Liang X., Wang G. A Convolutional Neural Network for Transportation Mode Detection Based on Smartphone Platform. IEEE 14th International Conference on Mobile Ad Hoc and Sensor Systems (MASS). 2017. Pp. 338–342. DOI: 10.1109/MASS.2017.81
13. Fang S., Fei Y., Xu Z. and Tsao Y. Learning Transportation Modes From Smartphone Sensors Based on Deep Neural Network. IEEE Sensors Journal. 2017. Vol. 17, No. 18. Pp. 6111–6118. DOI: 10.1109/JSEN.2017.2737825
14. Ana Alice Peregrino, Zhicheng Liu, Nivan Ferreira, Fabio Miranda. Transportation Scenario Planning with Graph Neural Networks. Conference: 10th International Workshop on Urban Computing. 2021. P. 1–5.
15. Pekel E., Selin Soner Kara. A comprehensive review for artificial neural network application to public transportation. Sigma: Journal of Engineering & Natural Sciences. 2017. Vol. 35 (1). P. 157–179.
16. Аулін В. В., Гриньків А. В., Головатий А. О., Лисенко С. В., Голуб Д. В., Кузик О. В., Тихий А. А. Методологічні основи проєктування та функціонування інтелектуальних транспортних і виробничих систем : монографія / під заг. ред. д. т. н., проф. Ауліна В. В. Кропивницький : Видавець Лисенко В. Ф., 2020. 428 с.
17. Amey Thakur, Archit Konde. Fundamentals of Neural Networks. International Journal for Research in Applied Science & Engineering Technology (IJRASET). 2021. Volume 9. Issue VIII. P. 407–426.
18. Хайкин С. Нейронные сети: полный курс, 2-е издание / Пер. с англ. Издательский дом Вильямс, 2006. 1104 с.
19. Терейковський І. А. Критерії вибору архітектури нейронної мережі для розв’язання задач із захисту інформації. Вісник Національного університету «Львівська політехніка». 2011. № 717. С. 168–177.
20. Pam Mantri, John Thomas. Design of Neural Networks. The 14th International Conference on Axiomatic Design (ICAD 2021). 2021. 1174 (2021). P. 1–16.
21. Савіна С. С., Бень В. П. Вибір архітектури нейромережі для розв’язання задачі класифікації надійності позичальників-фізичних осіб. Нейро-нечіткі технології моделювання в економіці. 2016. № 5. С. 124–151.
