The server-based automatic fare collection and management system in public transport and analysis with data mining techniques

Mass transit systems have existed for over a century. In the traditional payment structure; tickets, cards, and tokens are used for public transportation. Balances, discounts and transfer fee benefits are generally kept on cards specially designed for each city, and the cost of the ride is calculated and deducted by the validators while getting on the vehicle. The present thesis proposes a novel server-based fare collection, calculation and user management system. The proposed system provides personalized exclusive privileges to riders without getting the special card of the province and allows passengers to benefit from all the advantages of the cities' transport pricing. In addition, it removes the obstacles that prevent the use of new technologies such as contactless bank cards and Near Field Communication (NFC) standard like a traditional province card in public transport. The installation of the system with NFC usage was applied to several cities in Turkey. Real-world data was collected and then analyzed with two data mining techniques. First, the association rule mining technique was applied for the first time to find out NFC usage habits and user profile in public transportation. Second, classification algorithms were run on NFC-based transportation data and compared with each other in terms of accuracy rates. In order to show the personalization capability of the proposed system, this thesis also proposes a novel passenger scoring model, namely RFLT (Recency, Frequency, Loyalty, and Time), for offering a free pass promotion in public transportation. RFTL is the modified version of the well-known marketing method RFM (Recency, Frequency, Monetary). Experimental studies demonstrate the applicability of RFM model on transportation for the first time and show the comparison results of RFLT and wRFLT (weighted version) on a real-world dataset. The results of the study can be used to establish an efficient policy for increasing public transportation ridership.