DESIGN OF VEHICLE SPEED DETECTION ALGORITHM BASED ON MOBILE PHONE DATA IN SPATIAL PERSPECTIVE(CASE STUDY: ARTERIAL ROAD IN PADANG CITY)

• Main Author: Latersia Br Pinem, Enda
• Format: Theses
• Language: Indonesia
• Online Access: https://digilib.itb.ac.id/gdl/view/69032
• Institution: Institut Teknologi Bandung

One of the road maintenance activities is planning which begins with a vehicle speed survey. By knowing the vehicle speed characteristics, geometric planning elements such as gradients, superelevations, traffic intersections, and pavement thickness can be determined. Low speeds and long waiting times at a location on a road segment can cause the pavement to collapse. The study of vehicle speed becomes the basis for determining the basis for road maintenance budget policies. Utilization of mobile phone data for speed detection can provide an overview of speed spatially and temporally without the need for a field survey, thereby reducing costs and other resources in road management. The rapid increase in mobile phone usage has made available speed datasets that are easy to collect and update frequently for various applications. This study seeks to utilize mobile phone data to obtain vehicle speed information on the road. The data used are position, time, and device_id, as well as roads, and administrative boundaries. Mobile phone data recording from 06 to 12 January 2022 with the research location in the city of Padang. Analysis to obtain speed information is carried out by utilizing techniques in Geographic Information Systems (GIS) with classification, filtering, and sorting methods. The calculation results can show the maximum speed on the primary arterial road for each hourly interval. The maximum speed value indicates that there are motorized vehicles that can cross the road at that speed. However, this is not the case for secondary arterial roads, because there are still several secondary arterial roads that do not have speed values. This is because the coordinates, time, and device_id are collected only from the mobile phone in the vehicle that sends the current position on the road segment so that the information provided is only partially or when active using an application that shows the location. From this process, a mobile data processing algorithm was compiled based on continuous points in space and time which were ordered in observation to obtain vehicle speed. From this study, it can be concluded that vehicle speed detection using mobile phone data can be used to detect road speed if there are at least 2 continuous mobile phone position and time data. The results of this research are then compiled in an algorithm so that it can be easily applied and further developed.