سياوش طاهري نويد

Nowadays, the issue of big data and how to use them in various fields of engineering, such as applications in transportation planning, due to the rich resources of such data and their continuous recording and storing has become one of the important factors in analyzing and extracting parameters related to individual’s trip. Meanwhile, the rapid development of telecommunication networks and the increasing penetration rate of cell phones in recent years have provided valuable data sources on how people move among different places and times. This big data can be used in various fields of transportation, including the study of people's travel behavior, identification of movement patterns, impact studies, and etc. Extraction of Origin-Destination and Production-Attraction matrices are widely used examples of travel behavior that researchers have addressed. The advent of this data has changed the traditional process of data collection and processing, and usage of them, despite the disadvantages that any database may have, can make the results of transportation studies up-to-date and more reliable. In this research, an attempt has been made to estimate the Origin-Destination matrix and trip Production-Attraction between counties of Tehran province by home-based-work, home-based-other and non-home-based purposes, using recorded cellular data related to five days in Tehran province and using machine learning algorithms. Finally, these matrices were generalized to the whole community and compared for eva‎luation with the reported volumes of Iran Road Organization's traffic counts. The results show that the determination coefficient of R2 is 0.9738, the slope of regression is 0.9249, the Pearson coefficient is 0.9868 and the P-value 0.00085 is obtained, which indicates the good agreement of the estimated Origin-Destination matrix with the traffic counts data. Therefore, the estimated travel demand matrix can accurately estimate intercity trips in Tehran province.

ماتريس تقاضاي سفر , كلان داده , داده‌هاي تلفن همراه , استان تهران

Travel demand matrix , big data , mobile network data , Tehran Province

Siavash Taherinavid

Dr. Afshin Shariat Mohaymany