Embed Size (px)
Citation preview
Lin, Saat and Barkan – TRB 18-03877 1
Major Factors Affecting Passenger Train Accident Occurrence and Severity 1
in the United States: 1996 – 2015 2
3
Paper Number: 18-03877 4
5
Submitted for Presentation at the 97th Annual Meeting of the Transportation Research Board 6 and Publication in Transportation Research Record 7
8
Submission Date: 1 August 2017 9
10
Chen-Yu Lin1, Mohd Rapik Saat2, and Christopher P. L. Barkan 11
12
Rail Transportation and Engineering Center 13
Department of Civil and Environmental Engineering 14
University of Illinois at Urbana-Champaign 15
205 N. Mathews Ave., Urbana, IL, 61801 16
Fax: (217) 333-1924 17
18
19
Chen-Yu Lin
(217) 898-1841
Mohd Rapik Saat
(202) 639-2329
Christopher P. L. Barkan
(217) 244-6338
20
3,938 words + 8 Figures + 6 Tables = 7,438 Total words 21
1 Corresponding Author 2 Current Affiliation: Association of American Railroads. 425 Third St., SW, Washington, DC 20024
TRB 2018 Annual Meeting Original paper submittal - not revised by author.
Lin, Saat and Barkan – TRB 18-03877 2
Abstract 1
Demand for regional and intercity passenger transport in the United States is increasing, resulting 2 in the need to expand rail services and capacity. Railroads are viewed as a promising alternative 3 to highway and air transport because of their ability to provide safe, economical, comfortable, 4 and reliable transport. Passenger train safety has improved over the past two decades; however, 5 faster and more frequent service brings with it greater exposure to potential accidents. 6
Understanding risk in passenger train operations is essential to most efficient resource allocation 7 to further improve safety and reduce the risk of accidents and casualties. This paper presents an 8 analysis of passenger train accidents in the United States from 1996 to 2015 in order to 9 understand the general trend of passenger train accident rates, quantify the frequency and 10 severity of different accident types, and identify the major factors that cause them. This analysis 11
of train accident provides a foundation for further improvement in passenger train safety and 12
suggests opportunities for future research including shared-use rail corridor risk assessment, train 13 accident precursors, human factor analyses and data mining applications in railroad safety. 14
15
TRB 2018 Annual Meeting Original paper submittal - not revised by author.
Lin, Saat and Barkan – TRB 18-03877 3
Introduction 1
Demand for regional and intercity passenger transport in the United States is increasing, resulting 2 in the need to expand transportation network capacity. Railroads are viewed as a promising 3 alternative to highway and air transport because of their ability to provide safe, economical, 4 comfortable, and reliable transport. All types of passenger rail ridership have been growing in the 5 United States especially in recent years (Figure 1). Development of high-speed rail and higher-6
speed rail further highlight the increase in demand for faster and more frequent passenger rail 7 transportation (1, 2). 8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
TRB 2018 Annual Meeting Original paper submittal - not revised by author.
Lin, Saat and Barkan – TRB 18-03877 4
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
FIGURE 1 Passenger railroad ridership: (a) intercity (Amtrak), (b) commuter rail, (c) 29 transit and (d) light rail transit from 1996 – 2015 by million unlinked trips (3, 4) 30
(a)
Inte
rcity
(Am
tra
k)
(b)
Co
mm
ute
r R
ail
(c)
Tra
nsit
(d)
Lig
ht
Ra
il Tra
nsit
05
10
15
20
25
30
35
Ridership (Million Unlinked Trips)
0
10
0
20
0
30
0
40
0
50
0
60
0
Ridership (Million Unlinked Trips)
0
50
0
1,0
00
1,5
00
2,0
00
2,5
00
3,0
00
3,5
00
4,0
00
4,5
00
Ridership (Million Unlinked Trips)
0
10
0
20
0
30
0
40
0
50
0
60
0
Ridership (Million Unlinked Trips)
TRB 2018 Annual Meeting Original paper submittal - not revised by author.
Lin, Saat and Barkan – TRB 18-03877 5
As service and ridership grow, so do implications for safety, especially with higher speed 1 operation. A number of new or expanded services involve the use of shared trackage, rights-of-2 way (ROW) and corridors by passenger trains with freight or rail transit systems. The majority of 3
commuter and intercity passenger trains operate on or next to freight railroad corridors (5). 4 While these shared-use systems offer passenger rail operations benefits such as lower capital 5 costs, less environmental impact, and easier access to urban cores compared to building a new 6 dedicated line, they also incur some risk such as potential collisions between passenger trains 7 and freight trains (6) and other risk associated with greater exposure. Understanding these 8
potential aspects in passenger train operations is essential for better resource allocation to 9 improve passenger train safety and reduce the risk of accidents and casualties. 10
Literature Review 11
Research on train accident analyses in the United States has mostly focused on freight train 12
derailments (7-15), hazardous material releases (16-24) and grade crossing incidents (25-32). 13 Few studies have focused on quantitative analysis of U.S. passenger train safety. Much of the 14 research that has been done investigated passenger rail equipment crash energy management 15 systems intended to reduce casualties in a train collision or derailment (33-35). Lin et al. (36) 16
conducted a fault tree analysis to identify major factors that could lead to an adjacent track 17 accident on shared passenger and freight rail corridors and developed a semi-quantitative risk 18 assessment model to evaluate the risk (37). 19
Internationally, there have been more studies on passenger train accidents. Chen et al. used 20 Associated Rule and other data mining techniques to analyze Chinese passenger train accidents 21 (38). Ouyang et al. used System Theoretic Accident Models and Process (STAMP) to analyze a 22
severe railway accident in Jiaoji Railway in China (39). Niwa analyzed significant Japanese 23 railway accidents by five major aspects and conducted case studies of several severe accidents 24
(40). Evans used statistical methods to analyze fatal train accidents on Britain’s mainline 25 railways (41). The author proposed an exponential function to predict the declining trend of train 26
accident rates and applied to other mainline railway systems in Japan, Britain and Europe (42-27 44). Silla and Kallberg studied the development of the railway safety in Finland (45). Santos-28
Reyes and Beard used the Systemic Safety Management System (SSMS) model to analyze two 29 major passenger train accidents in the United Kingdom (46-47). Britton et al. conducted causal 30 analysis of train derailments in Australia (48). These studies are important in providing insights 31
for accident analysis methodologies and results for reference and comparison; however, there are 32 a number of differences in operating practices, rolling stock and organizational structure that 33
affect passenger train safety in the U.S. environment. Further study of passenger train accidents 34 is necessary to understand how to most effectively manage and reduce the risk of U.S. passenger 35 trains. 36
Research Objective 37
This paper presents an analysis of passenger train accidents in the United States from 1996 to 38 2015. The objective is to understand the general trend of passenger train accident rates, quantify 39 the frequency and severity of different accident types, and identify the major factors that cause 40 them. 41
42
TRB 2018 Annual Meeting Original paper submittal - not revised by author.
Lin, Saat and Barkan – TRB 18-03877 6
Passenger Train Accident Analysis 1996 – 2015 1
Train accident data from the United States Department of Transportation (U.S. DOT) Federal 2 Railroad Administration (FRA) was used for the analysis (49). Railroad accidents/incidents that 3 resulted in monetary loss exceeding a specific threshold must be reported to the Rail Equipment 4 Accident (REA) database maintained by FRA (50). This threshold is periodically adjusted for 5 inflation and is low enough so that only relatively minor incidents are not included. The FRA 6
categorizes train accidents into 13 types (Table 1). For the purpose of the analysis described in 7 this paper, these 13 types were consolidated into 5 groups. Incidents of defective pantograph or 8 overhead catenary system (OCS) do not represent the type of safety hazard of principal interest 9 in this research and were excluded. 10
11
TABLE 1 FRA Train Accident Type 12
13
14
Passenger train accident rate is calculated as number of accidents per million passenger train 15 miles. Over the past 20 years, the overall rate decreased from 0.99 accidents per million 16 passenger train miles in 1996, to 0.55 accidents per million passenger train miles in 2015 (Table 17 2). However, it is evident from Figure 2a that the decline fluctuated widely from year to year 18
especially during the first decade of the study period. The overall rate was broken down into the 19
five accident type groups defined above: derailment, collision, grade crossing, obstruction and 20
miscellaneous (Figure 2b). There was a fairly steady decline in derailments beginning in the mid-21 2000s. This is consistent with the decreasing trend in freight train derailments that occurred at 22 the same time (15). This reduction may be due, in part, to similar factors since most passenger 23 trains in the United States operate on freight railroad tracks. For other accident type groups, no 24 obvious trend was observed. Grade crossing accidents in general have the highest rate among all 25
train accident type groups and demonstrate no obvious downward trend, despite the substantial 26 general decline in grade crossing accidents (28, 51). 27
Accident/Incident Type Type Code Category in This Paper
Derailment 1 Derailment
Head-on collision 2 Collision
Rear collision 3 Collision
Side collision 4 Collision
Raking collision 5 Collision
Broken-train collision 6 Collision
Grade crossing incident 7 Grade Crossing
Railroad crossing collision 8 Collision
Obstruction 9 Obstruction
Explosive-detonation 10 Miscellaneous
Fire/violent rupture 11 Miscellaneous
Other impact 12 Miscellaneous
Others (with description) 13 Miscellaneous
Pantograph/OCS N/A Excluded
TRB 2018 Annual Meeting Original paper submittal - not revised by author.
Lin, Saat and Barkan – TRB 18-03877 7
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
TABLE 2 Annual Number of FRA-reportable Mainline Passenger Train Accident by 29 Accident Type, 1996 – 2015 30
Ye
ar
Ac
cid
en
tR
ate
Ac
cid
en
tR
ate
Ac
cid
en
tR
ate
Ac
cid
en
tR
ate
Ac
cid
en
tR
ate
Ac
cid
en
tR
ate
19
96
22
0.2
86
30
.03
93
00
.39
01
90
.24
72
0.0
26
76
0.9
87
19
97
16
0.2
05
10
.01
33
00
.38
41
40
.17
90
0.0
00
61
0.7
82
19
98
10
0.1
28
10
.01
31
30
.16
61
40
.17
91
0.0
13
39
0.4
98
19
99
60
.07
33
0.0
36
22
0.2
67
10
0.1
21
10
.01
24
20
.51
0
20
00
15
0.1
78
20
.02
42
40
.28
51
70
.20
22
0.0
24
60
0.7
12
20
01
13
0.1
48
60
.06
81
20
.13
71
60
.18
20
0.0
00
47
0.5
36
20
02
10
0.1
12
60
.06
72
00
.22
31
30
.14
50
0.0
00
49
0.5
47
20
03
15
0.1
68
00
.00
02
80
.31
31
70
.19
00
0.0
00
60
0.6
71
20
04
10
0.1
12
60
.06
75
30
.59
32
60
.29
11
0.0
11
96
1.0
75
20
05
11
0.1
22
30
.03
34
60
.51
22
80
.31
10
0.0
00
88
0.9
79
20
06
13
0.1
41
30
.03
33
30
.35
91
80
.19
60
0.0
00
67
0.7
29
20
07
16
0.1
68
40
.04
22
90
.30
42
20
.23
10
0.0
00
71
0.7
45
20
08
12
0.1
22
40
.04
13
70
.37
71
10
.11
20
0.0
00
64
0.6
53
20
09
11
0.1
07
30
.02
93
00
.29
12
30
.22
31
0.0
10
68
0.6
61
20
10
10
0.0
96
20
.01
93
30
.31
71
90
.18
20
0.0
00
64
0.6
15
20
11
10
0.0
92
30
.02
83
60
.33
02
00
.18
30
0.0
00
69
0.6
33
20
12
11
0.1
02
00
.00
03
10
.28
81
10
.10
20
0.0
00
53
0.4
93
20
13
10
0.0
90
20
.01
83
60
.32
51
40
.12
60
0.0
00
62
0.5
60
20
14
10
0.0
89
30
.02
73
10
.27
61
10
.09
80
0.0
00
55
0.4
90
20
15
70
.06
54
0.0
37
30
0.2
78
18
0.1
67
00
.00
05
90
.54
6
To
tal
23
80
.12
65
90
.03
16
04
0.3
20
34
10
.18
18
0.0
04
1,2
50
0.6
63
To
tal
De
railm
en
tC
ollis
ion
Gra
de
Cro
ss
ing
Ob
str
uc
tio
nM
isc
ella
ne
ou
s
TRB 2018 Annual Meeting Original paper submittal - not revised by author.
Lin, Saat and Barkan – TRB 18-03877 8
1
FIGURE 2 FRA-reportable mainline passenger train accident rates, 1996 – 2015 (a) overall 2 rates (b) rates by accident type 3
(a)
(b)
0.0
0.2
0.4
0.6
0.8
1.0
1.2A
ccid
en
t R
ate
(A
ccid
en
ts P
er
Mil
lio
nP
ass
en
ge
rT
rain
Mil
es)
Year
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
Accid
en
t R
ate
(A
ccid
en
ts P
er
Mil
lio
n
Pa
sse
ng
er
Tra
in M
ile
s)
Year
Derailment
Collision
Grade Crossing
Obstruction
Miscellaneous
TRB 2018 Annual Meeting Original paper submittal - not revised by author.
Lin, Saat and Barkan – TRB 18-03877 9
Risk is typically defined as the probability of a particular type of event multiplied by its severity 1 (52). In order to identify the types of accident that pose greater threat (i.e. higher probability, 2 severity, or both), accident rate and severity for each type of mainline passenger train accident 3
were plotted in a frequency-severity graph (Figure 3). Frequency-severity graphs are a helpful 4 risk visualization tool for train accidents because they enable comparison of the relative 5 frequency and severity of different accident types. They have been used in a number of other 6 railroad accident analyses (10, 13, 53). The graph is divided into four quadrants on the basis of 7 average frequency and severity along each axis. Frequency in this graph is defined as train 8
accident rate. Several different variables were considered to measure passenger train accident 9 severity. The number of railcars derailed has often been used as a proxy variable to measure 10 freight train accident severity (10, 13, 14, 17-19, 21, 22, 53). For passenger trains, casualties are 11 another pertinent variable to measure accident severity (37, 38, 40-44). In this paper, casualties 12 are defined as the total number of onboard passenger injuries and fatalities, and were used as the 13
primary severity indicator. 14
15
16
FIGURE 3 Frequency-severity graph for FRA-reportable mainline passenger train 17
accident, 1996 – 2015 18
19
Accident types in the upper right quadrant of Figure 3 are the most likely to pose the greatest risk 20 because they are both more frequent, and more severe, than average. None of the five accident 21 types fell in this quadrant, but derailments and collisions are most likely to result in high-casualty 22 incidents. Together, they accounted for about 24% of passenger train accidents, but caused about 23 65% of total casualties (Table 3). Derailments and collisions also cause more damage to rail 24
0
2
4
6
8
10
12
14
16
0.0 0.1 0.2 0.3 0.4
Ave
rag
e C
asu
alt
ies
Pe
r A
ccid
en
t
Accidents Per Million Passenger Train Miles
Average Frequency: 0.13
Average Casualties: 2.70
Collision
Derailment
MiscellaneousObstruction Grade Crossing
TRB 2018 Annual Meeting Original paper submittal - not revised by author.
Lin, Saat and Barkan – TRB 18-03877 10
equipment and infrastructure and are more likely to result in passenger and crew casualties. 1 Although grade crossing incidents are the most common type of accident, they are among the 2 least severe in their consequences to onboard passengers and crew. Therefore, the rest of this 3
paper examines mainline passenger derailments and collisions in more detail. 4
5
TABLE 3 Summary of Frequency, Accident Rate, Casualties and Average Casualties for 6 Different Types of Passenger Train Accidents, 1996 – 2015 7
8
9
Passenger Train Derailment and Collision Accident Cause Analysis 10
To further understand which factors contributed the most to passenger train derailments and 11
collisions, a more detailed causal analysis of the five accident types was conducted. FRA train-12 accident cause codes are hierarchically organized and categorized into major cause groups - 13
track, equipment, human factors, signal, and miscellaneous (50). Each of these major cause 14 groups has subgroups that include individual cause codes of related causes. In this paper, 15 alternative FRA subgroups developed by Arthur D. Little (ADL) and the Association of 16
American Railroads (AAR) are used in which similar cause codes were grouped based on 17
experts’ opinion (55). ADL’s groupings enable greater resolution for certain train accident 18 causes. For example, FRA combines broken rails, joint bars and rail anchors in the same 19 subgroup, whereas the ADL grouping distinguishes between broken rail and joint bar defects 20
(13). 21
The frequency and severity graph of accidents due to the major accident cause groups was 22 plotted (Figure 4). As in Figure 3, the graph is divided into four quadrants to enable comparison 23
of the relative frequency and severity of the different cause groups. The human factors cause 24 group had above-average frequency and was the most severe in terms of average casualties. It 25
accounted for 30% of the total derailments and collisions but 62.4% of the total casualties (Table 26 4). Track, Roadbed, and Structures accidents were more frequent than human factors, but less 27 severe (39.7% of the total derailments and collisions and 31.9% of the total casualties). Both 28
human factors and track, roadbed, and structure-related accident causes consistently represented 29 the most frequent and severe accident cause groups, together accounting for a total of 69.7% of 30 derailments and collisions, and 94.3% of casualties; therefore they were analyzed in more detail. 31
32
Frequency Percentage
Average
Accident Rate
Total
Casualties Percentage
Average
Casualties
Grade Crossing 604 48.3% 0.3202 1,050 30.3% 1.74
Obstruction 341 27.3% 0.1808 149 4.3% 0.44
Derailment 238 19.0% 0.1262 1,401 40.4% 5.89
Collision 59 4.7% 0.0313 864 24.9% 14.64
Miscellaneous 8 0.6% 0.0042 1 0.0% 0.13
Total 1,250 100.0% 0.1326 3,465 100.0% 2.77
TRB 2018 Annual Meeting Original paper submittal - not revised by author.
Lin, Saat and Barkan – TRB 18-03877 11
1
FIGURE 4 Frequency and severity graph of mainline passenger derailments and collisions, 2 1996-2015, by accident cause category with average casualties 3
4
5
TABLE 4 Summary of Frequency, Accident Rate, Casualties and Average Casualties of 6 Mainline Passenger Derailments and Collisions, 1996 – 2015, by Accident Cause Category 7
8
9
In order to identify trends in specific accident causes, the five-year moving average of combined 10 derailment and collision rate was broken down by accident cause category (Figure 5). 11
Infrastructure (Track, Roadbed and Structure) and human factor categories were consistently the 12 most frequent accident cause categories over the 20-year study period, with infrastructure causes 13 being higher for every 5-year interval except 1998 – 2002. 14
15
16
0
10
20
30
40
50
60
70
0 10 20 30 40
Av
era
ge
Ca
su
alt
ies
Pe
r A
cc
ide
nt
Number of Accidents
Train Operation Human Factors
Track, Roadbed and Structure
Miscellaneous
Mechanical and Electrical Factors
Signal and Communication
Average Frequency: 8.48
Average Casualties: 6.80
Failure to Obey/Display SignalsTrain Speed
Misc. Human Factors
Buckled Track
Non-Traffic, Weather Causes
Joint Bar Defects
Turnout Defects - SwitchesWide Gauge
Broken Rails or Welds
Track Geometry (excl. Wide Gauge)
Mainline Rules
Frequency Percentage
Average
Rate
Total
Casualties Percentage
Average
Casualties
Train Operation Human Factor 89 30.0% 0.0472 1,413 62.4% 15.88
Track, Roadbed, and Structure 118 39.7% 0.0626 723 31.9% 6.13
Miscellaneous 36 12.1% 0.0191 89 3.9% 2.47
Mechanical and Electrical Factors 49 16.5% 0.0260 40 1.8% 0.82
Signal and Communication 5 1.7% 0.0027 1 0.0% 0.20
Total 297 100.0% 0.1575 2,266 100.0% 7.63
TRB 2018 Annual Meeting Original paper submittal - not revised by author.
Lin, Saat and Barkan – TRB 18-03877 12
1
2
FIGURE 5 Five-year moving average of combined mainline passenger train derailment 3 and collision rate, 1996 – 2015, by accident cause category 4
5
The accident cause groups were further analyzed by preparing a frequency and severity 6 graph of the more detailed accident cause subgroups (Figure 6). Each data point represents one 7 accident cause subgroup. Data points with the same color and shape indicate that these accident 8
cause subgroups are in the same accident cause category. In terms of average casualties, four 9 accident cause subgroups were in the upper right quadrant, and thus most likely to pose the 10
greatest risk due to their high frequency and severity. All of them are human factor accident 11 causes: 12
Failure to Display/Obey Signals (Human Factors) 13 Train Speed (Human Factors) 14
Miscellaneous Human Factors (Human Factors) 15 Mainline Rules (Human Factors) 16
0.00
0.02
0.04
0.06
0.08
0.10
0.12
0.14
0.16
0.18
0.20
Ac
cid
en
t R
ate
(A
cc
ide
nts
Pe
r M
illio
n
Pa
ss
en
ger
Tra
in M
ile
s)
Five Year Moving Average Period
Total
Infrastructure
Human Factor
Equipment
Signal
Miscellaneous
TRB 2018 Annual Meeting Original paper submittal - not revised by author.
Lin, Saat and Barkan – TRB 18-03877 13
1
FIGURE 6 Frequency and severity graph of mainline passenger derailments and collisions, 2 1996-2015, by accident cause subgroups with average casualties 3
4
These four subgroups accounted for 20.5% of the total mainline passenger derailments 5 and collisions but 61.7% of total casualties (Table 5). Among all the subgroups identified in the 6 top-right quadrant, “Failure to Display/Obey Signals”, had the highest average casualties per 7
accident, followed by Mainline Rules, Train Speed and Misc. Human Factors. Irrespective of 8 quadrant, the five most frequent accident-cause subgroups were: Turnout Defect – Switches, 9
Failure to Obey/Display Signals, Wide Gauge, Use of Switches and Other Miscellaneous. 10 Combined they accounted for 43.8% of total derailments and collisions and 39.3% of total 11 casualties. Two of the top five most frequent accident cause subgroups were infrastructure 12 related, two of them were human factors and one of them was miscellaneous. 13
14
0
10
20
30
40
50
60
70
0 10 20 30 40
Av
era
ge
Ca
su
alt
ies
Pe
r A
cc
ide
nt
Number of Accidents
Train Operation Human Factors
Track, Roadbed and Structure
Miscellaneous
Mechanical and Electrical Factors
Signal and Communication
Average Frequency: 8.48
Average Casualties: 6.80
Failure to Obey/Display SignalsTrain Speed
Misc. Human Factors
Buckled Track
Non-Traffic, Weather Causes
Joint Bar Defects
Turnout Defects - SwitchesWide Gauge
Broken Rails or Welds
Track Geometry (excl. Wide Gauge)
Mainline Rules
TRB 2018 Annual Meeting Original paper submittal - not revised by author.
Lin, Saat and Barkan – TRB 18-03877 14
TABLE 5 Summary of Frequency, Accident Rate, Casualties and Average Casualties of 1 Mainline Passenger Derailments and Collisions, 1996 – 2015, by Accident Cause Subgroups 2
3
4
Number Percentage Number Percentage Average Number Percentage Average
10T Turnout Defects - Switches 38 12.8% 0.0201 84 3.7% 2.2 75 9.2% 2.0
05H Failure to Obey/Display Signals 29 9.8% 0.0154 701 31.2% 24.2 98 12.0% 3.4
03T Wide Gauge 23 7.7% 0.0122 32 1.4% 1.4 76 9.3% 3.3
11H Use of Switches 20 6.7% 0.0106 7 0.3% 0.4 34 4.2% 1.7
05M Other Miscellaneous 20 6.7% 0.0106 58 2.6% 2.9 66 8.1% 3.3
04T Track Geometry (excl. Wide Gauge) 19 6.4% 0.0101 48 2.1% 2.5 44 5.4% 2.3
15E Loco Trucks/Bearings/Wheels 13 4.4% 0.0069 29 1.3% 2.2 18 2.2% 1.4
08T Broken Rails or Welds 13 4.4% 0.0069 59 2.6% 4.5 56 6.9% 4.3
08H Mainline Rules 11 3.7% 0.0058 106 4.7% 9.6 16 2.0% 1.5
10H Train Speed 11 3.7% 0.0058 247 11.0% 22.5 32 3.9% 2.9
01M Obstructions 11 3.7% 0.0058 10 0.4% 0.9 41 5.0% 3.7
13E Other Wheel Defects (Car) 10 3.4% 0.0053 5 0.2% 0.5 14 1.7% 1.4
12H Misc. Human Factors 10 3.4% 0.0053 331 14.7% 33.1 35 4.3% 3.5
18E All Other Car Defects 9 3.0% 0.0048 5 0.2% 0.6 10 1.2% 1.1
12T Misc. Track and Structure Defects 8 2.7% 0.0042 3 0.1% 0.4 19 2.3% 2.4
06E Centerplate/Carbody Defects (Car) 5 1.7% 0.0027 1 0.0% 0.2 2 0.2% 0.4
01S Signal Failures 5 1.7% 0.0027 1 0.0% 0.2 8 1.0% 1.6
11E Other Axle/Journal Defects (Car) 4 1.3% 0.0021 0 0.0% 0.0 10 1.2% 2.5
02H Handbrake Operations 4 1.3% 0.0021 7 0.3% 1.8 14 1.7% 3.5
17E All Other Locomotive Defects 3 1.0% 0.0016 0 0.0% 0.0 5 0.6% 1.7
07H Switching Rules 3 1.0% 0.0016 0 0.0% 0.0 3 0.4% 1.0
03M Lading Problems 3 1.0% 0.0016 0 0.0% 0.0 0 0.0% 0.0
02T Non-Traffic, Weather Causes 3 1.0% 0.0016 173 7.7% 57.7 21 2.6% 7.0
05T Buckled Track 3 1.0% 0.0016 144 6.4% 48.0 25 3.1% 8.3
06T Rail Defects at Bolted Joint 3 1.0% 0.0016 30 1.3% 10.0 20 2.5% 6.7
07T Joint Bar Defects 3 1.0% 0.0016 124 5.5% 41.3 22 2.7% 7.3
14E TOFC/COFC Defects 2 0.7% 0.0011 0 0.0% 0.0 6 0.7% 3.0
04M Track-Train Interaction 2 0.7% 0.0011 1 0.0% 0.5 5 0.6% 2.5
01T Roadbed Defects 2 0.7% 0.0011 15 0.7% 7.5 8 1.0% 4.0
09T Other Rail and Joint Defects 2 0.7% 0.0011 10 0.4% 5.0 4 0.5% 2.0
07E Coupler Defects (Car) 1 0.3% 0.0005 0 0.0% 0.0 1 0.1% 1.0
09E Sidebearing, Suspension Defects (Car) 1 0.3% 0.0005 0 0.0% 0.0 2 0.2% 2.0
19E Stiff Truck (Car) 1 0.3% 0.0005 0 0.0% 0.0 1 0.1% 1.0
04H Employee Physical Condition 1 0.3% 0.0005 14 0.6% 14.0 18 2.2% 18.0
11T Turnout Defects - Frogs 1 0.3% 0.0005 1 0.0% 1.0 5 0.6% 5.0
01E Air Hose Defect (Car) 0 0.0% 0.0000 0 0.0% 0.0 0 0.0% 0.0
02E Brake Rigging Defect (Car) 0 0.0% 0.0000 0 0.0% 0.0 0 0.0% 0.0
03E Handbrake Defects (Car) 0 0.0% 0.0000 0 0.0% 0.0 0 0.0% 0.0
04E UDE (Car or Loco) 0 0.0% 0.0000 0 0.0% 0.0 0 0.0% 0.0
05E Other Brake Defect (Car) 0 0.0% 0.0000 0 0.0% 0.0 0 0.0% 0.0
08E Truck Structure Defects (Car) 0 0.0% 0.0000 0 0.0% 0.0 0 0.0% 0.0
10E Bearing Failure (Car) 0 0.0% 0.0000 0 0.0% 0.0 0 0.0% 0.0
12E Broken Wheels (Car) 0 0.0% 0.0000 0 0.0% 0.0 0 0.0% 0.0
16E Loco Electrical and Fires 0 0.0% 0.0000 0 0.0% 0.0 0 0.0% 0.0
20E Track/Train Interaction (Hunting) (Car) 0 0.0% 0.0000 0 0.0% 0.0 0 0.0% 0.0
21E Current Collection Equipment (Loco) 0 0.0% 0.0000 0 0.0% 0.0 0 0.0% 0.0
01H Brake Operation (Main Line) 0 0.0% 0.0000 0 0.0% 0.0 0 0.0% 0.0
03H Brake Operations (Other) 0 0.0% 0.0000 0 0.0% 0.0 0 0.0% 0.0
06H Radio Communications Error 0 0.0% 0.0000 0 0.0% 0.0 0 0.0% 0.0
09H Train Handling (excl. Brakes) 0 0.0% 0.0000 0 0.0% 0.0 0 0.0% 0.0
02M Grade Crossing Collisions 0 0.0% 0.0000 0 0.0% 0.0 0 0.0% 0.0
Total 297 100.0% 0.027 2,246 100.0% 7.6 814 100.0% 2.7
Cause
Subgroup
Code
Frequency Cars Derailed
Cause Subgroup Description
Accidents
Per Million
Train MilesCasualties
TRB 2018 Annual Meeting Original paper submittal - not revised by author.
Lin, Saat and Barkan – TRB 18-03877 15
Effect of Speed on Passenger Train Derailment and Collision Cause 1 Previous research has found that the speed of a train at the time of derailment was positively 2 correlated with with derailment severity (10, 12-14, 16-18, 20). Previous research has also found 3
an inverse relationship between FRA track class and freight train derailment rate (11, 14, 16). 4
Figure 7 shows the number and percentage of mainline passenger train derailments and collisions 5 by speed and accident cause category. The majority of train accidents – about 57% – occurred 6
below 20 mph. This may be related to the high incidence of defective-turnout-caused 7 derailments. Turnouts are found at stations, terminals, and the ends of sidings where trains are 8 likely to slow down due to speed restrictions, scheduled stops or meet/pass activities. 9 Infrastructure related accidents occurred in almost all speed ranges and had the highest 10 percentage except the 101 mph+ category. No specific trends were found for human-factor-11
caused and equipment-caused accidents; however, the three accidents that occurred with train 12
speed greater than 100 mph were human factor and equipment caused. 13
TRB 2018 Annual Meeting Original paper submittal - not revised by author.
Lin, Saat and Barkan – TRB 18-03877 16
1
FIGURE 7 Number (a) and percentage (b) of mainline passenger train derailments and 2 collisions by speed and accident cause category, 1996 – 2015 3
(a)
(b)
0
10
20
30
40
50
60
70
80
0-20 21-40 41-60 61-80 81-100 101+
Nu
mb
er
of A
cc
ide
nt W
ith
in S
pe
ed
Ra
ng
e
Accident Speed (mph)
Infrastructure
Human Factor
Equipment
Signal
Miscellaneous
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
0-20 21-40 41-60 61-80 81-100 101+
Pe
rce
nta
ge
of A
cc
ide
nt W
ith
in S
pe
ed
Ra
ng
e
Accident Speed (mph)
Infrastructure
Human Factor
Equipment
Signal
Miscellaneous
TRB 2018 Annual Meeting Original paper submittal - not revised by author.
Lin, Saat and Barkan – TRB 18-03877 17
To further understand what caused a derailment or collision at different speeds, Table 6 shows 1 the number of mainline passenger train derailments and collisions by accident cause subgroup in 2 different speed ranges. In the 0-20 mph range, Turnout Defects – Switches was the top accident 3
cause subgroup, consistent with our previous inference regarding low speed accident causes. In 4 the 21-40 mph and 41-60 mph ranges, Failure to Obey/Display Signals was most frequent. In the 5 61-80 and 80-100 ranges, some equipment-related accident cause subgroups, namely All Other 6 Car Defects and Other Wheel Defects (Car) were the top causes. The three accidents in which 7 speed was above 100 mph were in the following three subgroups: All Other Locomotive Defects, 8
Centerplate/Carbody Defect (Car) and Misc. Human Factors. Summaries of these three accidents 9 are as follows (in the order of accident date): 10
1. April 12th, 2001. Amtrak train was side-swiped by an improperly secured freight 11
locomotive door from a freight train on the adjacent track (accident type: raking collision; 12
Amtrak train speed: 110 mph, no cars derailed; no casualties; accident cause sub group: 13 All Other Locomotive Defects) 14
2. January 24th, 2004. Amtrak train was side-swiped by an improperly secured freight car 15
door from a freight train on the adjacent track (accident type: raking collision; Amtrak 16 train speed: 110 mph, no cars derailed, no casualties, accident cause subgroup: 17 Centerplate/Carbody Defect (Car)) 18
3. May 12th, 2015. Amtrak train derailed at 102 mph, resulting in 1 locomotive and 7 19 passenger cars derailed; 229 casualties; accident cause subgroup: Misc. Human Factors 20
(56) 21
22
TABLE 6 Most Frequent Accident Cause Subgroups of Mainline Passenger Train 23 Derailment and Collision, 1996 – 2015, by Accident Speed Ranges 24
25
26
A scatterplot of train accident speed versus casualties was prepared (Figure 8). A simple linear 27 regression indicates a positive relationship between train accident speed and casualties (P-value 28
0-20 21-40 41-60 61-80 81-100 100+*
1Turnout Defects -
Switches
Failure to
Obey/Display
Signals
Failure to
Obey/Display
Signals
All Other Car
Defects
Other Wheel
Defects (Car)
Misc. Human
Factors
2
Failure to
Obey/Display
Signals
Loco
Trucks/Bearings
/Wheels
Broken Rails or
WeldsJoint Bar Defects
Track Geometry
(excl. Wide
Gauge)
All Other
Locomotive
Defects
3 Use of Switches Wide Gauge Wide GaugeBroken Rails or
WeldsTrain Speed
Centerplate/Carb
ody Defects (Car)
4 Wide GaugeTurnout Defects -
Switches
Other
Miscellaneous
Misc. Human
Factors
Handbrake
Operations
5Other
MiscellaneousMainline Rules Obstructions
Track Geometry
(excl. Wide
Gauge)
Non-Traffic,
Weather Causes
* There were only three accidents where train speed was greater than 100 mph.
Accident Speed (mph)
Top
Accident
Cause
Subgroups
Ranking by
Frequency
TRB 2018 Annual Meeting Original paper submittal - not revised by author.
Lin, Saat and Barkan – TRB 18-03877 18
= 0.038; R2 = 0.11); however, the coefficient of determination is low, meaning there are other 1 factors also affecting severity. One example is type of accident. A raking collision between an 2 object from a freight train and the side of a passenger train may not result in large number of 3
casualties, even at high speed (17 occurrences; average train speed = 55.5 mph; average 4 casualties = 0.94). On the other hand, a head-on collision may result in severe casualties at 5 relative lower speed (6 occurrences; average train speed = 19.4 mph; average casualties = 54.9). 6 Other factors include number of passengers onboard, curvature and grade of the track and 7 location of accident such as on a bridge or in a tunnel. A positive relationship between train 8
accident speed and casualties is reasonable because when an accident occurs at higher speed, 9 more kinetic energy is involved and therefore there is a higher probability of greater 10 consequences. 11
12
13
FIGURE 8 Scatterplot of casualties and train accident speed 14
15
DISCUSSION 16
In this paper we analyze U.S. mainline passenger train accidents in the 20-year period from 1996 17 to 2015 and identify major accident types and causes. The paper also associated the effect of 18 train speed with accident frequency, severity and accident causes. These findings provide a basis 19 for future improvement in passenger train safety. Based on these results, several directions for 20
future research are discussed. 21
Adjacent Track Accidents on Shared-Use Rail Corridors 22
0
50
100
150
200
250
0 20 40 60 80 100 120
Ca
su
alt
ies
Train Accident Speed (mph)
TRB 2018 Annual Meeting Original paper submittal - not revised by author.
Lin, Saat and Barkan – TRB 18-03877 19
With the development of high-speed rail as well as continuing improvement in the conventional 1 passenger rail system in the United States, it is expected that there will be more shared-use rail 2 corridors and more mixed passenger and freight train operations. A consequent safety issue on 3
these shared-use rail corridors is adjacent track accident risk (6). Adjacent track accidents are a 4 type of accident in which one train derails and intrudes upon adjacent tracks, and then is struck 5 by trains on those tracks. With more trains operating on a corridor, the probability of train 6 interactions also increases, meaning that if a train derails and intrudes onto an adjacent track, 7 there is a greater chance that another train will be present or approaching on the adjacent track. 8
Furthermore, higher speed of train operations on shared-use rail corridors means the potential 9 consequences of an accident are also greater. Focusing on addressing the risk of adjacent track 10 accidents will help improve our understanding of this risk and lead to more effective risk 11 reduction strategies. 12
Accident Precursors 13
As safety continues to improve in the railroad system, statistical analyses to reliably estimate risk 14 will become more challenging due to the smaller empirical basis for analysis (52). To address 15 this, accident precursors must be considered. An “Accident Precursor” is defined by National 16
Aeronautics and Space Administration (NASA) as: 17
“an anomaly that signals the potential for more severe consequences that may occur in 18 the future, due to causes that are discernible from its occurrence today (57).” 19
An example of an accident precursor in a railroad system is locomotive engineer over-running a 20
stop signal, but without any further consequence such as a collision or derailment. Train 21 accidents are a subset of accident precursors, meaning that under certain conditions, accident 22 precursors will result in train accidents, but most will not. Analyzing accident precursors 23
provides more data and consequently more robust predictive risk analysis. Studying accident 24 precursors also allows researchers to identify preventive measures that mitigate risk at the 25
precursor event level, and potentially further reduce the occurrence of train accidents. For 26 example, if a preventive measure can effectively reduce the probability of an engineer 27
overrunning a stop signal, it can also reduce the probability of a train accident caused by Failure 28 to Obey/Display Signal. Positive Train Control for passenger train operation is an example of a 29 preventive measure that will prevent this type of precursor event, as well as accidents associated 30
with this cause, and certain others as well. 31
Accident precursor analysis has been implemented in the United Kingdom (58), and in the 32 United States. The Confidential Close Call Reporting System has been implemented by the FRA 33 to collect close call (or “near misses”) data (59) and these data are a good candidate for train 34
accident precursor analysis. 35
Human Factors Analysis 36
Train Operation Human Factors were identified as the most frequent and severe passenger train 37 accident cause category. Consequently addressing them will be critical to the success of further 38 passenger train risk reduction efforts. Railroad human factor research encompasses a wide 39
spectrum of topics including human fatigue in train operation, ergonomics, and human 40 performance in the train control system. In Europe, Signal Passed at Danger (SPAD), which is 41 similar to the Failure to Obey/Display Signal cause subgroup in the United States, is a critical 42
TRB 2018 Annual Meeting Original paper submittal - not revised by author.
Lin, Saat and Barkan – TRB 18-03877 20
human factor topic and has drawn substantial attention as an important source of the risk (60, 1 61). Employee fatigue is another aspect of human factors risk and has contributed to several 2 severe passenger train accidents. Overall, there are opportunities and potential to reduce 3
passenger train derailments and collision by addressing human factors. 4
Data Mining Applications in Railroad Safety Improvement 5
Expanded automated data collection systems, combined with rapid advances in data mining 6 technology, means that new methodologies are available for rail safety analyses. These include 7 associated rules (38), STAMP (39), and Maximal Information Coefficient (MIC) (62). Data 8 mining techniques can be implemented to increase risk model accuracy and to handle complex 9 effects from multiple (and perhaps correlated) influencing factors. 10
11
CONCLUSION 12 This paper presents the results of a study to identify the most important factors contributing to 13
the risk of passenger train accidents. Derailments and collisions were identified as the most 14 potentially significant train accident types, while human factors accidents and track failures were 15
the primary causes of those accidents. Accident causes related to human factors and train 16 operations such as train speed violations and failure to obey signals have high risk. High-risk 17 infrastructure-related factors include track geometry defects and broken rails or welds. Most 18
passenger train derailments and collisions occurred at lower speed. This analysis of train accident 19 causes is important for rational allocation of resources to reduce accident occurrence and 20
consequences and provides a foundation for further improvement in passenger train safety. The 21 paper also suggests opportunities for future research including shared-use rail corridor risk 22 assessment, train accident precursors, human factor analyses and data mining applications in 23
railroad safety. 24
25 ACKNOWLEDGEMENTS 26 This research was supported by the National University Rail (NURail) Center, a U.S. DOT OST-27 R University Transportation Center. The views expressed in this paper do not necessarily reflect 28
the views of the Association of American Railroads, the second author’s current employer. 29
30 31 REFERENCES 32
1. Federal Railroad Administration. High Speed Rail Strategic Plan. United States 33 Department of Transportation, Washington D.C., 2009. 34
2. Federal Railroad Administration. High Speed Intercity Passenger Rail (HSIPR) Program. 35
https://www.fra.dot.gov/Page/P0134. Accessed June 2017. 36
3. American Public Transportation Association. Ridership Report Archives. 37 http://www.apta.com/resources/statistics/Pages/RidershipArchives.aspx. Accessed 38 November 2016. 39
4. National Railroad Passenger Cooperation. Amtrak National Facts. 40 http://www.amtrak.com/servlet/ContentServer?c=Page&pagename=am%2FLayout&cid=41 1246041980246. Accessed December 2016. 42
TRB 2018 Annual Meeting Original paper submittal - not revised by author.
Lin, Saat and Barkan – TRB 18-03877 21
5. Brod, D., Metcalf, A.E. and Kraft, E. Web-Based Screening Tool for Shared-Use Rail 1 Corridors. Publication NCFRP Report No. 27. Transportation Research Board of the 2 National Academies, Washington, D.C., 2014. 3
6. Saat, M.R. and Barkan, C.P.L. Investigating Technical Challenges and Research Needs 4 Related to Shared Corridors for High-Speed Passenger and Railroad Freight 5 Operations. Publication DOT/FRA/ORD-13-29. United States Department of 6 Transportation, Washington, D.C., 2013. 7
7. Birk, A.M., Anderson, R.J. and Coppens, A.J. A computer simulation of a derailment 8
accident: Parts I – model basis. Journal of Hazardous Materials, Vol. 25, 1990, pp. 121 – 9 147. 10
8. Birk, A.M., Anderson, R.J. and Coppens, A.J. A computer simulation of a derailment 11 accident: Parts II – sample simulation. Journal of Hazardous Materials, 1990, Vol. 25, 12
pp. 149 – 165. 13
9. Dennis, S.M. Changes in railroad track accident rates. Transportation Quarterly, Vol. 56, 14
Issue 4, 2002, pp 161 – 174. 15
10. Barkan, C.P.L., Dick, C.T. and Anderson, R.T. Railroad derailment factors affecting 16 hazardous materials transportation risk. Transportation Research Record: Journal of 17
Transportation Research Board, Vol. 1825, 2003, pp. 64 – 74. 18
11. Anderson, R.T. and Barkan, C.P.L. Railroad accident rates for use in transportation risk 19
analysis. Transportation Research Record: Journal of Transportation Research Board, 20 Vol. 1863, 2004, pp. 88 – 98. 21
12. Liu, X., Barkan, C.P.L. and Saat, M.R. Analysis of derailments by accident cause: 22
evaluating railroad track upgrades to reduce transportation risk. Transportation Research 23
Record: Journal of Transportation Research Board, Vol. 2261, 2011, pp. 178 – 185. 24
13. Liu, X., Saat, M.R. and Barkan, C.P.L. Analysis of causes of major train derailment and 25 their effect on accident rates. Transportation Research Record: Journal of 26
Transportation Research Board, Vol. 2289, 2012, pp. 154 – 163. 27
14. Liu, X., Saat, M.R. and Barkan, C.P.L. Analysis of U.S. freight-train derailment severity 28
using zero-truncated negative binomial regression and quantile regression. Accident 29 Analysis & Prevention, Vol. 59, 2013, pp. 87 – 93. 30
15. Liu, X. Statistical temporal analysis of freight train derailments rates in the United States: 31 2000 to 2012. Transportation Research Record: Journal of Transportation Research 32 Board, Vol. 2476, 2015, pp. 119 – 125. 33
16. Nayak, P.R., Rosenfield, D.B. and Hagopian, J.H. Event Probabilities and Impact Zones 34 for Hazardous Materials Accidents on Railroads. Publication DOT/FRA/ORD-83/20. 35
United States Department of Transportation, Washington, D.C., 1983. 36
17. Saccomanno, F.F. and El-Hage, S.M. Minimizing derailments of railcars carrying 37 dangerous commodities through effective marshaling strategies. Transportation Research 38 Record: Journal of Transportation Research Board, Vol. 1245, 1989, pp. 34 – 51. 39
TRB 2018 Annual Meeting Original paper submittal - not revised by author.
Lin, Saat and Barkan – TRB 18-03877 22
18. Saccomanno, F.F. and El-Hage, S.M. Establishing derailment profile by position for 1 corridor shipment of dangerous goods. Canadian Journal of Civil Engineering, Vol. 18, 2 Issue 1, 1991, pp. 67 – 75. 3
19. Kawprasert, A. and Barkan, C.P.L. Effect of route rationalization on hazardous materials 4 transportation risk. Transportation Research Record: Journal of Transportation Research 5 Board, Vol. 2043, 2008, pp. 65 – 72. 6
20. Bagheri, M., Saccomanno, F.F., Chenouri, S. and Fu, L. Reducing the threat of in-transit 7 derailments involving dangerous goods through effective placement along the train 8
consist. Accident Analysis & Prevention, Vol. 43, Issue 3, 2011, pp. 613 – 620. 9
21. Liu, X., Saat, M.R. and Barkan, C.P.L. Integrated risk reduction framework to improve 10 railway hazardous materials transportation safety. Journal of Hazardous Materials, Vol. 11 260, 2013, pp. 131 – 140. 12
22. Liu, X., Saat, M.R. and Barkan, C.P.L. Probability analysis of multiple-tank-car release 13 incidents in railway hazardous materials transportation. Journal of Hazardous Materials, 14
Vol. 276, 2014, pp. 442 – 451. 15
23. Liu, X. and Hong, Y. Analysis of railroad tank car releases using a generalized binomial 16 model. Accident Analysis & Prevention, Vol. 84, 2015, pp. 20 – 26. 17
24. Liu, X. Optimizing rail defect inspection frequency to reduce the risk of hazardous 18 materials transportation by rail. Journal of Loss Prevention in the Process Industries, 19
Vol. 48, 2017, pp. 151 – 161. 20
25. Benekohal, R.F. and Elzohairy, Y.M. A new formula for prioritizing railroad crossings 21 for safety improvement. In: Proceedings of the Institution of Transportation Engineers 22
Annual Meeting, Chicago, IL, 2001. 23
26. Austin, R.D. and Carson, J. L. An alternative accident prediction model for highway-rail 24 interfaces. Accident Analysis & Prevention, Vol. 34, Issue 1, 2002, pp. 31 – 42. 25
27. Saccomanno, F.F., Fu, L. and Miranda-Moreno, L.F. Risk-based model for identifying 26
highway-rail grade crossing blackspots. Transportation Research Record: Journal of 27 Transportation Research Board, Vol. 1862, 2004, pp. 127 – 135. 28
28. Mok, S.C. and Savage, I. Why has safety improved at rail-highway grade crossings? Risk 29 Analysis, Vol. 25, Issue 4, 2005, pp. 867 – 881. 30
29. Saccomanno, F.F., Park, P.Y.J. and Fu, L. Estimating countermeasure effects for 31 reducing collisions at highway-rail grade crossings. Accident Analysis & Prevention, Vol. 32 39, Issue 2, 2007, pp. 406 – 416. 33
30. Chadwick, S. G., Zhou, N. and Saat, M.R. Highway-rail grade crossing safety challenges 34 for shared operations of high-speed passenger and heavy freight in the U.S. Safety 35
Science, Vol. 68, 2014, pp. 128 – 137. 36
31. Williams, T., Abello, J., Betak, J. and Desimore, D. Using data visualization to analyze 37 grade crossing accidents. In Proceedings of the 2015 Joint Rail Conference, San Jose, 38 CA, 2015. 39
TRB 2018 Annual Meeting Original paper submittal - not revised by author.
Lin, Saat and Barkan – TRB 18-03877 23
32. Williams, T., Nelson, C. and Betak, J. Applying topic modeling to railroad grade crossing 1 accident report text. In Proceedings of the 2015 Joint Rail Conference, San Jose, CA, 2 2015. 3
33. Tyrell, D.C. US rail equipment crashworthiness standards. Proceedings of the Institution 4 of Mechanical Engineers, Part F: Journal of Rail and Rapid Transit, Volume 216, Issue 5 2, 2002, pp. 123 – 130. 6
34. Tyrell, D.C. Rail passenger equipment accidents and the evaluation of crashworthiness 7 strategies. Proceedings of the Institution of Mechanical Engineers, Part F: Journal of 8
Rail and Rapid Transit, Volume 216, Issue 2, 2002, pp. 131 – 147. 9
35. Tyrell, D.C. and Perlman, A.B. Evaluation of rail passenger equipment crashworthiness 10 strategies. Transportation Research Record: Journal of Transportation Research Board, 11 Vol. 1825, 2003, pp. 8 – 14. 12
36. Lin, C.Y., Saat, M.R. and Barkan, C.P.L. Fault tree analysis of adjacent track accidents 13 on shared-use rail corridors. Transportation Research Record: Journal of Transportation 14
Research Board, Vol. 2546, 2016, pp. 129 – 136. 15
37. Lin, C.Y. and Saat, M.R. Semi-quantitative risk assessment of adjacent track accidents on 16 shared-use rail corridors. In: Proceedings of the 2014 Joint Rail Conferences, Colorado 17
Springs, CO, 2014. 18
38. Chen, D., Xu, C. and Ni, S. Data mining on Chinese train accidents to derive associated 19
rules. Proceedings of the Institution of Mechanical Engineers, Part F: Journal of Rail 20 and Rapid Transit, Volume 231, Issue 2, 2017, pp. 239 – 252. 21
39. Ouyang, M., Hong, L., Yu, M.H. and Fei, Q. STAMP-based analysis on the railway 22
accident and accident spreading: Taking the China-Jiaoji railway accident for example. 23
Safety Science, Vol. 48, Issue 5, 2010, pp. 544 – 555. 24
40. Niwa, Y. A proposal for a new accident analysis method and its application to a 25 catastrophic railway accident in Japan. Cognition, Technology & Work, Vol. 11, Issue 3, 26
2009, pp. 187 – 204. 27
41. Evans, A.W. Fatal train accidents on Britain’s mainline railway. Journal of the Royal 28
Statistical Society. Series A (Statistics in Society), Vol. 163, No 1, 2000, pp. 99 – 119. 29
42. Evans, A.W. Rail safety and rail privatisation in Britain. Accident Analysis & Prevention, 30 Vol. 39, Issue 3, 2007, pp. 510 – 523. 31
43. Evans, A.W. Rail safety and rail privatisation in Japan. Accident Analysis & Prevention, 32 Vol. 42, Issue 4, 2010, pp. 1296 – 1301. 33
44. Evans, A.W. Fatal train accidents on Europe’s railways: 1980-2009. Accident Analysis & 34 Prevention, Vol. 43, Issue 1, 2011, pp. 391 – 401. 35
45. Silla, A. and Kallberg, V.P. The development of railway safety in Finland. Accident 36 Analysis & Prevention, Vol. 45, 2012, pp. 737 – 744. 37
46. Santos-Reyes, J. and Beard, A.N. A systemic analysis of the Paddington railway accident. 38 Proceedings of the Institution of Mechanical Engineers, Part F: Journal of Rail and 39 Rapid Transit, Volume 220, Issue 2, 2006, pp. 121 – 151. 40
TRB 2018 Annual Meeting Original paper submittal - not revised by author.
Lin, Saat and Barkan – TRB 18-03877 24
47. Santos-Reyes, J. and Beard, A.N. A systemic analysis of the Edge Hill railway accident. 1 Accident Analysis & Prevention, Vol. 41, Issue 6, 2009, pp. 1133 – 1144. 2
48. Britton, M.A., Asnaashari, S. and Read, G.J.M. Analysis of train derailment cause and 3
outcome in Victoria, Australia, between 2007 and 2013: Implications for regulation. 4 Journal of Transportation Safety & Security, Vol. 9, Issue 1, 2017, pp. 45 – 63. 5
49. Federal Railroad Administration. Download Data on Demand. U.S. Department of 6 Transportation, Washington, D.C. 7 http://safetydata.fra.dot.gov/OfficeofSafety/default.aspx. Accessed July 1st, 2017. 8
50. Federal Railroad Administration. FRA Guide for Preparing Accident/Incident Reports. 9 U.S. Department of Transportation, Washington, D.C., 2011. 10
51. Federal Railroad Administration. In-Depth Data Analysis of Grade Crossing Accidents 11
Resulting in Injuries and Fatalities. Publication DOT/FRA/ORD-17-04. United States 12 Department of Transportation, Washington, D.C., 2017. 13
52. Elvik, R. and Voll, N.G. Challenges of improving safety in very safe transport systems. 14
Safety Sciences, Vol. 63, 2014, pp. 115 – 123. 15
53. Wang, B. Z., Barkan, C.P.L., Liu, X. and Saat, M.R. Trends in U.S. freight train accident 16 causes and rates – A quantitative approach (working paper). 17
54. Schafer II, D.H. and Barkan, C.P.L. Relationship between train length and accident 18 causes and rates. Transportation Research Record: Journal of Transportation Research 19
Board, Vol. 2043, 2008, pp. 73 – 82. 20
55. Arthur D. Little, Inc. (ADL). Risk Assessment for the Transportation of Hazardous 21 Materials by Rail, Supplementary Report: Railroad Accident Rate and Risk Reduction 22
Option Effectiveness Analysis and Data (Second Revision). Cambridge, Massachusetts, 23
1996. 24
56. Derailment of Amtrak Passenger Train 188 Philadelphia, Pennsylvania May 12, 2015. 25 National Transportation Safety Board. 26
https://www.ntsb.gov/investigations/AccidentReports/Reports/RAR1602.pdf Accessed 27 June 29, 2017. 28
57. NASA Accident Precursor Analysis Handbook. National Aeronautics and Space 29 Administration. 30 https://permanent.access.gpo.gov/gpo27108/20120003292_2012003430.pdf Accessed 31 June 29, 2017. 32
58. Prospective Study into Harmonised Train Accident Precursors Analysis and 33
Management. The European Railway Agency. http://www.era.europa.eu/Document-34 Register/Documents/PPR665%20Report.pdf Accessed June 29, 2017. 35
59. Close Call Reporting. Federal Railroad Administration. https://www.fra.dot.gov/c3rs 36 Accessed June 29, 2017. 37
60. Pasquini, A., Rizzo, A. and Save., L. A methodology for the analysis of SPAD. Safety 38 Science, Vol. 42, Issue 5, 2004, pp.437 – 455. 39
TRB 2018 Annual Meeting Original paper submittal - not revised by author.
Lin, Saat and Barkan – TRB 18-03877 25
61. Wilson, J.R. and Norris, B.J. Rail human factors: Past, present and future. Applied 1 Ergonomics, Vol. 36, Issue. 6, 2005, pp. 649 – 660. 2
62. Shao, F. and Li, K. A graph model for preventing railway accidents based on the 3
maximal information coefficient. International Journal of Modern Physics B, Vol. 31, 4 Issue 3, 2017, 1750010 (19 pages). 5
TRB 2018 Annual Meeting Original paper submittal - not revised by author.
