Parking lot availability prediction and patterns - 2

• Main Author: Yeo, Ying Xiang
• Other Authors: Ho Shen-Shyang
• Format: Final Year Project
• Language: English
• Published: 2015
• Subjects: DRNTU::Engineering::Computer science and engineering::Software
• Online Access: http://hdl.handle.net/10356/62889
• Institution: Nanyang Technological University

Searching for a vacant parking lot in Singapore during peak periods is a huge challenge that results in time and money wasted. With the use of cloud-based intelligent car parking services, such problems can be solved by incorporating parking strategy that utilizes the existing information available on the web. Also driver can carry out better decision making after receives vital or relevant information from the system when finding the best optimal car park. Therefore the purpose of this project is to provide car parking services that help resolve or reduce the problems encountered in the process of finding an optimal parking lot. A parking recommendation system was developed to assist drivers to find the best suitable parking location. The system takes into consideration of driver preferences based on factors such as driving distance, walking distance and parking cost. Also a cost calculator system was developed to provide an estimated cost for parking in specified car parks to the users. The system takes into account attributes such as the rules of the car parking fee, duration, day and time for parking car. Also, it provides a way for users to perform a cost comparison when choosing a preferred car park among a list of candidate car parks. The cost calculator system was able to generate the cost based on the real data containing the parking rule. The real dataset for rules are collected from the Land Transport Authority (LTA) Data Mall. These data provided by LTA are given in XML format. Both systems are developed as backend web services in PHP and Java programming language. Also the recommendation system was able to provide a list of best recommended car park based on ranking scores which derive from factors mentioned previously.