INTELLIGENCE DEVELOPMENT ON INTERNET OF THINGS-BASED SMART PARKING SYSTEM TO OPTIMIZE PARKING MANAGEMENT IN CAMPUS PARKING AREAS (STUDY CASE : BANDUNG INSTITUTE OF TECHNOLOGY)

• Main Author: Ilyas Irfan Syiraaj, Muhammad
• Format: Final Project
• Language: Indonesia
• Online Access: https://digilib.itb.ac.id/gdl/view/73936
• Institution: Institut Teknologi Bandung

The condition of lectures during the post-COVID-19 pandemic has made lecture activities begin to be carried out offline so that activities on campus have increased, including the activity of parking vehicles in the campus parking area. The large number of students who use vehicles to go to campus makes the campus parking lot always full of students' vehicles. Most campus parking areas do not have an area large enough to accommodate all the vehicles owned by students, for example, the Bandung Institute of Technology (ITB) campus parking area. Students who use campus parking at ITB encounter several problems that often occur. These problems will be solved by utilizing technology in the form of computer vision and machine learning with specific problems including lack of tracking of vehicle identity, lack of parking quota information and students unable to get out of the parking lot because there are motorcycles that are parked in an untidy manner. The solution proposed for this problem is an intelligent parking system that utilizes computer vision and machine learning to predict parking quotas, detect vehicles, and detect vehicle parking neatness. The method used to develop an intelligent parking system is the system engineering method written by Alexander Kossiakoff combined with the general method of making computer vision models and machine learning. This final project research will use systems engineering methods, but the system will only be developed up to the engineering development stage. The results of the final assignment research are in the form of an intelligent parking system to increase the efficiency of the existing campus parking system with a functional testing success percentage of 100% and a performance testing success percentage of 95%.