Open AccessPrediction of traffic flows by applying the statistical method
Author(s) -
Ol'ga Lebedeva,
J. O. Poltavskaya
Publication year - 2020
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/1680/1/012032
Subject(s) - reliability (semiconductor) , traffic flow (computer networking) , flow (mathematics) , environmental science , names of the days of the week , process (computing) , econometrics , statistics , computer science , mathematics , power (physics) , linguistics , physics , geometry , computer security , philosophy , quantum mechanics , operating system
Traffic flows in the streets network undergo changes not only during the day, but also on the weekdays. Existing models take these changes into account as unpredictable fluctuations. Flows may vary systematically due to differences in the schedules of industrial enterprises, warehouses, and terminals. By analyzing the statistics, weekdays or seasons are determined when the flows predictably differ. To process the data, it is necessary to develop models or amend the initial data of the model (in the case of everyday dynamic models) to verify the reliability of the proposed methods. In order to investigate daily variability, as well as take into account the dynamics during the day, the distribution of flows is analyzed using functional linear models. The result of the study is the ability to predict traffic flow by day of the week or season. Testing is carried out on real data. The daily circulation was found to vary significantly for each day of the week, including differences in time and peak flux duration, as well as the relationship between peak and peak peaks. This technique aims to increase the efficiency and reliability of the transport sector as a whole.