Choice Model for Leisure Vehicle Considering Park & Ride on Subway: A Case of Xianlin University City in Nanjing

Abstract:

Tow kinds of leisure vehicles are considered here, private car or parking & riding on subway. A case of Xianlin University City in Nanjing is chosen to help train and test a BP neural network, based on which multivariate linear regression assists to research the way of how to raise the use of parking & riding on subway. A four-layered structure, composed of those influencing factors, is designed first for establishing a neuron model which will help choose from driving or parking & riding on subway. This structure includes 6 indexes and 11 neurons. Then, a BP neural network trained and tested with the data from the case of the Xianlin University in Nanjing, consists of 3 layers with 15 nodes in the middle with the error less than 10^-3. The original data are gained by Delphi technique, questionnaire, measurement and calculation. Finally, some conclusions are gained. Its considered necessary to continue improving the parking conditions around leisure places, especially for those health and amusement ones. Its also important to focus on the exit locations of subway stations, the route of underpass and the inner circumstance there. The real distance for foot to the destination is not so crucial as thought before, while the perceptible one plays a role. Its a good way to inform those private cars owners the conditions of the transportation congestions before their decisions on the leisure vehicles.

• Periodical: Applied Mechanics and Materials (Volumes 361-363)
• Pages: 2146-2151
• DOI: https://doi.org/10.4028/www.scientific.net/AMM.361-363.2146
• Online since: August 2013
• Authors: Juan Zhang*
• Keywords: BP Neural Network (BPNN), Leisure Vehicle, Multivariate Linear Regression, Park Ride on Subway, Xianlin University City

References

[1] Aolfazlm, Eric JM, Nested logic models and artificial neural networks for predicting household automobile choice comparison of performance, Transportation Research Record (2002), 92-100.

DOI: 10.3141/1807-12

[2] Davidah, Tutt, A comparison of the predictive potential of artificial neural networks and nested logit models for commuter mode choice, Transportation Research Part E, 36 (2000), 155-172.

DOI: 10.1016/s1366-5545(99)00030-7

[3] Jianchuan Xianyu, Zhicai Juan, Research of travel mode choice with knowledge-based neural network, Application Research of Computers, 25(2008), 2651-2654. (In Chinese).

[4] Hua Yu, Xiaolin Zhang, Feifei Huang etc., Analysis on urbanity development differentiations of county areas in Jiangsu province aased on BP neural network, Human Geography, 24(2009), 38-49. (In Chinese).

[5] Zhenfang Huang, Ye Zhu, Linwang Yuan, The intention, classification and evaluation of leisure tourism resources: A case study of Changzhou, Jiangsu Province, Geographical Research, 30(2011), pp.1543-1553.

[6] Defeng Zhang, MATLAB Nerve Net Design, first ed., China Machine Press, Beijing, 2009. (In Chinese).

[7] Dingyu Xue, Yangquan Chen, Solving Advanced Applied Mathematical Problems Via MATLAB, first ed., Tsinghua University Press, Beijing, 2008. (In Chinese).

[8] Martin T. Hagan, Howard B. Demuth, Mark H. Beale, Neural Network Design, first ed., China Machine Press, Beijing, 2002. (In Chinese).