Smart Card Data Mining of Public Transport Destination: A Literature Review

The urban population in 2014 accounted for 54% of the total global population, up from 34% in 1960, and continues to grow. The global urban population is expected to grow approximately 1.84%, 1.63% and 1.44% between 2015 and 2020, 2020 and 2025, and 2025 and 2030 respectively. This growing population puts pressure on government not only to accommodate the current and potential citizens but also provide them facilities and services for a better living standard. Providing a sustainable growing environment for the citizens is the biggest challenge for the government. As the populations increase, complexity network of transportation, water and sanitation, emergency services, etc. will increase many folds. SMART CITY Mission is being implemented to resolve this issue. As the cities turn smart, so should the transportation facilities. India on June 2018 had only 20 cities with populations of over 500,000 have organized public transport systems, pointing to the large gap to be bridged in their journey to turn smart. The aim of this paper is to examine the impact of smart card data from public transport for improving the predictions and planning of public transport usage and congestions. The mobile apps like M-Indicator, Google Maps don’t interlink, do not have a real time tracking of vehicles, fare distribution, congestion-based route mapping for public transportation. These factors are addressed in the paper with its advantages and disadvantages. This paper also talks about how information from smart card is to be extracted, how Big Data is to be managed and finally come to a smart, sustainable Urban Transit System. This paper also brings into light the data security issues and measures to curb those issues. This paper proposes and emphasizes on a single smart card for all modes of public transport

Download Full-text

Improving Population Demand Estimation with Transit Chaining Breaks

The Open Transportation Journal ◽

10.2174/1874447801812010319 ◽

2018 ◽

Vol 12 (1) ◽

pp. 319-325

Author(s):

Jin Haitao ◽

Jin Fengjun ◽

Ni Yong ◽

Huang Jianling ◽

Du Yong

Keyword(s):

Data Mining ◽

Public Transport ◽

Public Transportation ◽

Population Distribution ◽

Correlation Coefficients ◽

Demand Estimation ◽

New Approach ◽

Pearson Product Moment Correlation ◽

Transportation Demand ◽

Smart Card Data

Background: Data mining of smart card data collected through AFC systems have proved useful in estimations of public transport demand. Whereas most estimations of demand are made by analyzing transit orientations or destinations of unchained transits. However, organization of bus or metro routes compels riders to make a lot of unnecessary transfers, and the transfer points are neither reflective of population’s actual orientations nor of their destinations. Aims and Objectives: The objective of this paper is to improve estimations of population demand by identifying transfer activities of riders using public transportation. Durations and displacements of transit chaining breaks are to be check in judging the transfer activities. Boarding stops for making transfers are ruled out as transportation demand estimation. The effectiveness of the new approach entailing the use of transit chaining breaks is also to be evaluated based on the calculation of Pearson product-moment correlation coefficients for assessing the correlation between transportation estimation and population distribution. Result and Conclusion: Durations and displacements of transit chaining breaks could be used to identify transfer activities. The use of the transit chaining approach reduces the occurrence of false demand, resulting in the estimation being more objective in relation to the population. The results of the study indicated that the inclusion of transit chaining breaks leads to more accurate estimations of public transport demand within a population.

Download Full-text