Analyzing Large-Scale Smart Card Data to Investigate Public Transport Travel Behaviour Using Big Data Analytics

Abstract

In urban public transport, Smart card data have been used more and more in order to collect fare automatically. They allowed passengers to access almost all type of public transportation system modes (bus, train, tram, funiculars, LRT, metro, and ferryboats) with a single card that is valid for the complete journey. Although Smart card major concentration is in revenue collection, they also generate massive amounts of passive data from the technological devices installed to control the operation of them. Generated data could be beneficial to transit planners which could rise the better understanding of passengers' behavioral patterns for short and long term service planning. However, one of the major challenges is the fact that traditional infrastructures and methods are inefficient when processing or analyzing a large volume of data. Thus, as an alternative, big data technology could be employed to enhance collecting, storing, processing, and analyzing the data. Moreover, the main motivation would be cost-efficiency of this methodology as the cost of processing and analyzing large-scale data is huge. This experience demonstrates that a combination of planning knowledge, big data, and data mining tool allows to produce travel behaviors indicators, public transport policies, operational performance, and fare policies.

Ekstraksi Perilaku Komuter Pada Commuter Line Menggunakan Rule-Based Machine Learning

“…Sistem AFC memungkinkan management mendapatkan data terperinci dalam jumlah besar mengenai rute setiap komuter setiap hari. Management kereta Commuter Line dapat menganalisis data tersebut untuk mendapatkan informasi tetang perilaku perjalanan dan perpindahan komuter dalam jumlah besar yang memudahkah operator kereta Commuter Line melakukan evaluasi kualitas layanan dan mengoptimalkan strategi operasional kereta commuter [2], [3].…”

Using Smart Farecard Data to Support Transit Network Restructuring: Findings from Los Angeles

“…The emergence of smart farecard data, which record both geographic and temporal footprints over time, has made it feasible to conduct longitudinal analysis on transit users ( 4, 5 ). Recognized as a legitimate source of “big data,” there is a growing body of literature that delves into various dimensions of transit planning—travel demand variability ( 6, 7 ), transfer behavior ( 8, 9 ), and heuristics—to impute transit origin–destination (O-D) matrices ( 10–13 ). Smart farecard data can inform several levels of planning—strategic and long-term (user behavior, customer segmentation, demand analysis), tactical (schedule adjustments, load profiles, network restructuring), and operational (evaluation of day-to-day performance indicators on reliability, productivity, etc.)…”