||Analysis of Rail Line Trespassing Accidents-
A Case Study of the TRA System
||Department of Transportation & Communication Management Science
Rail line trespassing accident
Count data regression model
TOBIT regression model
Geographically weighted regression model
Rail transport, characterized by its exclusive right of way and high service capacity, has received increasing attention in the international transport community. According to the statistics of Taiwan Railways Administration (TRA), the TRA served approximately 232.8 million passengers in 2014 with an average daily volume of 637,881 passengers, an increase of 4.99% compared to that of 2013. The constant increase of passenger trips has indicated that TRA plays a crucial role for intercity transport services in Taiwan. In terms of traffic accidents for the TRA system, it is relatively few compared to that of the highway system. Nevertheless, in case of a rail accident, it usually causes significant number of casualties and different degrees of impact, even induces some external costs (such as train delay, locomotive engineer’s psychological impact, and injury recovery and facility maintenance expenses, etc.). Past studies concerning rail accidents mainly focused on highway-rail grade crossing accidents. Investigation of rail trespassing incidents is relatively rare. Currently, the operation mileage of the TRA system is 1061.3 kilometers including a total of 91 kilometers of the branch lines. Along the main and branch lines of the TRA system, there are only 466 grade crossings with different control means (mainly active control by quadrant-type gate). Compared to those traffic accidents occurred on grade crossings, the accident frequency and potential impacts of rail line trespassing incidents along the entire service scope of the TRA system should not be ignored, and this topic is deserved in-depth investigation.
To resolve the above research question, this study aims to investigate the casual relationship between rail trespassing incidents and a set of covariates using a most recent ten-year (2005-2014) railway line incident dataset provided by the TRA by a two-stage research approach. In the first stage of this study, rail trespassing accident frequency data are modeled by Poisson regression, Negative binomial regression, TOBIT regression models. In the second stage, local traffic and environmental characteristics are incorporated into the modeling process. Geographically weighted regression model is developed to further evaluate the effects of local characteristics on the occurrence of a rail trespassing incident. According to the empirical results of this study, the local model has a better goodness of fit and is able to explore the heterogeneity in different regions. Thereby, the empirical study results of the local models are expected to provide the TRA and related government offices with a reference in preparing rail line trespassing incident prevention strategies.
TABLE OF CONTENTS V
LIST OF FIGURES VII
LIST OF TABLES VIII
Chapter 1 Introduction 1
Chapter 2 Background and Literature Review 9
2.1 Rail Trespassing Incidents 9
2.2 Statistical Models for Rail Incidents and/or accidents 12
2.2.1 Count Data Regression Model 13
2.2.2 Tobit Regression Model (TRM) 13
2.2.3 Geographically Weighted Regression Model (GWR) 14
2.3 Summary 14
Chapter 3 Methodology 16
3.1 Count Data Regression Model 18
3.1.1 Poisson Regression Model (PRM) 18
3.1.2 Negative Binomial Regression Model (NBRM) 20
3.2 Tobit Regression Model (TRM) 24
3.3 Geographically Weighted Regression (GWR) 26
3.3.1 Weighted functions 28
3.3.2 The bandwidth calculation 29
3.3.3 Fixed Kernel or Adaptive Kernel 30
3.4 Summary 31
Chapter 4 Empirical Study 33
4.1 Data collection and description 33
4.2 Descriptive Statistics 37
4.3 Results and Discussion 41
4.4 Global Model 41
4.5 Local Model 45
4.6 Discussion 47
Chapter 5 Conclusion and Recommendation 57
5.1 Findings and conclusions 57
5.2 Future research and recommendations 58
Appendix A: Descriptive Statistics for Trespassing Variables 67
Appendix B: Original data of rail line trespassing accidents provided by TRA 71
Appendix C: Rearranged data of rail line trespassing accidents prepared by this study 77
1.Anastasopoulos, P. C., Tarko, A. P., and Mannering, F. L. (2008). Tobit analysis of vehicle accident rates on interstate highways. Accident Analysis and Prevention, 40(2), 768-775.
2.Anastasopoulos, P. C., and Mannering, F. L. (2011). An empirical assessment of fixed and random parameter logit models using crash-and non-crash-specific injury data. Accident Analysis and Prevention, 43(3), 1140-1147.
3.Anastasopoulos, P. C., Mannering, F. L., Shankar, V. N., and Haddock, J. E. (2012a). A study of factors affecting highway accident rates using the random-parameters tobit model. Accident Analysis and Prevention, 45, 628-633.
4.Anastasopoulos, P. C., Shankar, V. N., Haddock, J. E., and Mannering, F. L. (2012b). A multivariate tobit analysis of highway accident-injury-severity rates. Accident Analysis and Prevention, 45, 110-119.
5.Annual Safety Performance Report. (2012). Rail Safety and Standards Board, London.
6.Apple Daily Post. (2016) Woman was sent to the police office because of taking pictures on the tracks. Retrieved on April 10, 2016. http://www.appledaily.com.tw/realtimenews/article/local/20160410/835570/
7.Chen, C.B. (2007). Applying Adaptive Cruise Control System to Preventing Rear-End Collisions on Highway. (Unpublished Work). National Cheng Kung University.
8.Chi, P.F. (1997). A Study on Accident Analysis Models for Sun Yat-Sen National Freeway (Unpublished Work). National Chiao Tung University.
9.Chinanews.com, (2012). The most dangerous of Thailand "Railway market". http://big5.chinanews.com/tp/hd2011/2012/08-18/125738.shtml
10.Chiou, Y. C., Jou, R. C., and Yang, C. H. (2015). Factors affecting public transportation usage rate: Geographically weighted regression. Transportation Research Part A: Policy and Practice, 78, 161-177.
11.Chiou, Y.C., Weng, C.H., Chen, W.Y., Wang, M.H., Fu, C., Tseng, H.M., and Lin, Y.S. (2014). Modeling Mode Choice Behaviors by Using Mixed Logit Models. Transportation Planning Journal, 43(2), 143-171.
12.Evans, A. W. (2003). Accidental fatalities in transport. Journal of the Royal Statistical Society: Series A (Statistics in Society), 166(2), 253-260.
13.Federal Railway Administration Office of Safety Analysis. (2012). http://safetydata.fra.dot.gov/OfficeofSafety/publicsite/query/AccidentByRegionStateCounty.aspx.
14.Fotheringham, A. S., Brunsdon, C., and Charlton, M. (2000). Quantitative geography: perspectives on spatial data analysis. Sage.
15.Fotheringham, A. S., Brunsdon, C., and Charlton, M. (2003). Geographically weighted regression: The analysis of spatially varying relationships. John Wiley & Sons.
16.Fotheringham, S., Charlton, M., and Brunsdon, C. (1998). Geographically weighted regression: A natural evolution of the expansion method for spatial data analysis. Environment and planning A, 30(11), 1905-1927.
17.Hadayeghi, A., Shalaby, A. S., and Persaud, B. N. (2010). Development of planning level transportation safety tools using geographically weighted Poisson regression. Accident Analysis and Prevention, 42(2), 676-688.
18.Harrison, J. A. (2015). 2015 ROW Fatality & Trespass Prevention Workshop. Research Results, 15-42.
19.Havârneanu, G. M., Burkhardt, J. M., and Paran, F. (2015). A systematic review of the literature on safety measures to prevent railway suicides and trespassing accidents. Accident Analysis and Prevention, 81, 30-50.
20.Hilbe, J. M. (2009). Logistic regression models. CRC Press.
21.Hsiao, Y.P. (2011). A Study on Potential Factors influencing Passenger Car Accident in Taipei Metropolitan Area. (Unpublished Work). National Chiao Tung University.
22.Hu, S. R., Li, C. S., and Lee, C. K. (2010). Investigation of key factors for accident severity at railroad grade crossings by using a logit model. Safety science, 48(2), 186-194.
23.Hu, S.R., and Ji, J.L. (2014). Investigating Risk Factors of Traffic Accidents at Highway-Railroad Grade Crossings Using Counted Data Statistical Modeling: A Case Study of the TRA System. Transportation Planning Journal, 43(2), 115-142.
24.Huang, C.F. (2009). The analysis of dangerous degrees for railway level crossing accidents. (Unpublished Work). National Central University.
25.Huang, J.H. (2013). Relationship between Driver Violation and Traffic Accidents. (Unpublished Work). National Chiao Tung University.
26.Huang, Y.U. (2007). A Spatial Analysis of Factors Affecting Housing Price in Taipei– Geographically Weighted Regression Approach. (Unpublished Work). National Taipei University.
27.Huang. W.S. (2005). Exploring the Factors Affecting the Level Crossing Accidents of TRA. (Unpublished Work). National Chiao Tung University.
28.Institute of Transportation, TALAS. Retrieved on May 10, 2016. http://talas-pub.iot.gov.tw/Default.aspx.
29.Kim, J. K., Ulfarsson, G. F., Shankar, V. N., and Mannering, F. L. (2010). A note on modeling pedestrian-injury severity in motor-vehicle crashes with the mixed logit model. Accident Analysis and Prevention, 42(6), 1751-1758.
30.Kyriakidis, M., Hirsch, R., and Majumdar, A. (2012). Metro railway safety: An analysis of accident precursors. Safety science, 50(7), 1535-1548.
31.Liberty Times Net. TRA spent 430 million setting fencing alone 175 kilometers to reduce casualties. Retrieved on March 6, 2016. http://m.ltn.com.tw/news/focus/paper/965150, retrived on June 6, 2016?
32.Lo, S.P. (2002). The Research of Accident Serious of the Taiwan Railway Administration. (Unpublished Work). National Cheng Kung University.
33.Lobb, B. (2006). Trespassing on the tracks: A review of railway pedestrian safety research. Journal of safety research, 37(4), 359-365.
34.Lobb, B., Harre, N., & Suddendorf, T. (2001). An evaluation of a suburban railway pedestrian crossing safety programme. Accident Analysis & Prevention, 33(2), 157-165.
35.Lord, D., and Mannering, F. (2010). The statistical analysis of crash-frequency data: a review and assessment of methodological alternatives. Transportation Research Part A: Policy and Practice, 44(5), 291-305.
36.McFadden, D., and Train, K. (2000). Mixed MNL models for discrete response. Journal of applied Econometrics, 15(5), 447-470.
37.Milton, J. C., Shankar, V. N., and Mannering, F. L. (2008). Highway accident severities and the mixed logit model: an exploratory empirical analysis. Accident Analysis & Prevention, 40(1), 260-266.
38.Ministry of Land, Transport and Tourism. Retrieved from http://www.mlit.go.jp/toukeijouhou/chojou/stat-e.htm
39.Morse, D. (2015). Taking photos on train tracks: ‘A mistake you can’t undo.’ The Washington Post. https://www.washingtonpost.com/local/public-safety/taking-photos-on-train-tracks-a-mistake-you-cant-undo/2015/10/02/99769f94-6218-11e5-9757-e49273f05f65_story.html.
40.Pai, C. W., Hwang, K. P., and Saleh, W. (2009). A mixed logit analysis of motorists’ right-of-way violation in motorcycle accidents at priority T-junctions. Accident Analysis & Prevention, 41(3), 565-573.
41.Savage, I. (2007). Trespassing on the Railroad. Research in Transportation Economics, 20, 199-224.
42.Silla, A., and Luoma, J. (2011). Effect of three countermeasures against the illegal crossing of railway tracks. Accident Analysis & Prevention, 43(3), 1089-1094.
43.Silla, A., and Luoma, J. (2012a). Main characteristics of train–pedestrian fatalities on Finnish railroads. Accident Analysis and Prevention, 45, 61-66.
44.Silla, A., and Luoma, J. (2012b). Opinions on railway trespassing of people living close to a railway line. Safety science, 50(1), 62-67.
45.Smith, A. (2006). Level crossing safety performance report June 2006. Rail Safety and Standards Board, London.
46.Stanchak, K., and daSilva, M. (2014). Trespass Event Risk Factors (No. DOT/FRA/ORD-14/32). Federal Railroad Administration. http://www.fra.dot.gov/eLib/details/L16047#p1_z50_gD_lRE_lRO_lRT_ktrespassing
47.Taiwan Railways Administration. Retrieved on January 1, 2015. http://www.railway.gov.tw/tw/index.html
48.Transportation Safety Board of Canada. Retrieved on May 10, 2016. http://www.tsb.gc.ca/eng/rail/index.asp
49.Tsai, W.J. (2013). The Analysis of Measurements and Influence Factors of Mixed Land Use. (Unpublished Work). National Cheng Kung University.
50.Wang, J.C. and Guo, Z.G. (2004). Logistic regression models: Methods and Applications. Wunan Publishing Co.
51.Wang, X., Liu, J., Khattak, A. J., and Clarke, D. (2016). Non-crossing rail-trespassing crashes in the past decade: A spatial approach to analyzing injury severity. Safety Science, 82, 44-55.
52.Wang, Y.D. (2006). The Study of Performance Evaluation for Financial Holding Corporations— Publicly Listed Companies on Taiwan Stock Exchange and Over-the-Counter Market (Unpublished Work). Ming Chuan University.
53.Washington, S. P., Karlaftis, M. G., & Mannering, F. (2003). Statistical and econometric methods for transportation data analysis. CRC press.
54.Wen, C.H. (2009). Theory and Applications of Discrete Choice Models. Journal of National Defense Management, 30(2), 1-15.
55.Xie, Y.D. (2010). Analysis of Traffic Accident and Safety Improvement Measures for Unsignalized Intersection in Taipei City. (Unpublished Work). National Chiao Tung University.
56.Yang, C.H. (2012). Determinants of Public Transportation Usage Rate. (Unpublished Work). National Chiao Tung University.
57.Yang, H.P. (2011). An empirical study of the demand for railway station: geographically weighted regression approach. (Unpublished Work). National Taiwan University.
58.Yang, K.P. (2007). Accident Frequency and Impact Analysis of Railway Level Crossing. (Unpublished Work). National Cheng Kung University.
59.Yuan, S.M. (2003). An Application of the Mixed Multinomial Logit Model on Taiwan Housing Choice (Master's thesis). National Cheng Kung University.