U.S. Railroad Safety Statistics
and Trends
Peter W. French
AVP- Safety & Performance Analysis
Association of American Railroads
July 29, 2008
Railroad Safety: Topics
• Safety Statistics & Trends
–
–
–
–
–
–
Train Safety (Train Accidents)
Grade Crossing Safety
Trespassers
Passenger Safety
Employee Safety
Hazardous Materials Safety
Railroad Safety: Topics
(continued)
•
•
•
•
RR Initiatives to Improve Safety
Improvements in Technology
Safety Programs
Risk-Based Safety Performance Standards
U.S. Railroad Safety Statistics:
Main Themes
• Railroads have dramatically improved safety over the
last two and a half decades.
• Railroads compare favorably with other industries &
transportation modes.
• The most troubling railroad safety problems arise
from factors largely outside railroad control.
• Railroads have implemented numerous and effective
technological improvements and company-wide safety
programs.
U.S. Railroad Safety Statistics:
In 2007, U.S. railroads achieved their safest year
ever by the following major safety yardsticks:
• Train Accident Rate, down 10% from 2006.
• Grade Crossing Collisions and Collision Rates,
down 6% and 4% from 2006.
Employee Casualties and Rates missed the 2006
records by 0.8% and 1.4%.
Sources: FRA website: http://safetydata.fra.dot.gov/Prelim/2007/r01.htm (Preliminary 2007 data)
FRA, Railroad Safety Statistics Annual Report, 1997-2005, Tables 1-1, 1-2.
FRA, Accident/Incident Bulletin, 1975-1996, Tables 1, 51.
In 2007, 95% of rail-related fatalities were
grade crossing users and trespassers.
Trespassers: 473
(56%)
Grade Crossing
Users: 337
(39%)
Passengers: 4
Employees: 17
Others: 20
Sources: FRA website: http://safetydata.fra.dot.gov/Prelim/2007/r03.htm (preliminary 2007 data).
FRA, Railroad Safety Statistics Annual Report 2005, Table 1-3.
In 2007, 60% of rail-related injuries were incurred
by employees on duty.
Trespassers:
Passengers: 1,058
398
Grade Crossing
Users: 828
Others: 1,254
Employees 5,263
Sources: FRA website: http://safetydata.fra.dot.gov/Prelim/2007/r03.htm (preliminary 2007 data).
FRA, Railroad Safety Statistics Annual Report 2005, Table 1-3.
Train accidents per million train-miles have dropped
71% since 1980 and 30% since 1990, to a new low.
12
11.43
10
8
6
4.73
3.54
4
3.29
2
0
1980
1985
1990
1995
2000
2007
Sources: FRA website: http://safetydata.fra.dot.gov/Prelim/2007/r01.htm (preliminary 2007 data).
FRA, Railroad Safety Statistics Annual Report, 1997-2005, Tables 1-1, 1-2.
FRA, Accident/Incident Bulletin, 1980-1996, Tables 19, 36.
Note: Excludes grade crossing accidents.
Train collisions per million train-miles have dropped
85% since 1980 and 51% since 1990.
2.0
1.67
1.5
1.0
0.52
0.5
0.25
0.0
1980
1985
1990
1995
2000
2007
Sources: FRA website: http://safetydata.fra.dot.gov/Prelim/2007/r01.htm (preliminary 2007 data).
FRA, Railroad Safety Statistics Annual Report, 1997-2005, Tables 1-1, 5-6.
FRA, Accident/Incident Bulletin, 1980-1996, Tables 19, 36.
Note: Excludes grade crossing collisions.
Mainline train collisions per million train-miles on Class I freight
railroads have dropped 85% since 1980 and 54% since 1990.
0.6
0.54
0.52
All Collisions: All RRs
0.4
0.37
0.2
Mainline & Sidings:
Class I Freight RRs
All Consists
0.17
Frt Trains Only
0.254
0.12
0.080
0.057
0.0
1980
1985
1990
1995
2000
2007
Sources: FRA website: http://safetydata.fra.dot.gov/Prelim/2007/r01.htm. AAR Analysis of FRA train
accident database through 2007. FRA, Railroad Safety Statistics Annual Report, 1997-2005,
Tables 1-1, 5-6; FRA, Accident/Incident Bulletin, 1980-1996, Tables 19, 36.
Note: Excludes grade crossing collisions. Includes passenger train collisions on Class I freight railroads.
Derailments per million train-miles have dropped
74% since 1980 and 33% since 1990, to a new low.
10
8.98
8
6
3.52
4
2.38
2
0
1980
1985
1990
1995
2000
2007
Sources: FRA website: http://safetydata.fra.dot.gov/Prelim/2007/r01.htm & r02.htm (preliminary 2007 data)
FRA, Railroad Safety Statistics Annual Report, 1997-2005, Tables 1-1, 5-6.
FRA, Accident/Incident Bulletin, 1980-1996, Tables 19, 36.
Note: Excludes grade crossing accidents.
Equipment-caused accidents per million train-miles
have dropped 80% since 1980 and 42% since 1990.
2.5
2.03
2.0
1.5
1.0
0.70
0.5
0.40
0.0
1980
1985
1990
1995
2000
2007
Sources: FRA website: http://safetydata.fra.dot.gov/Prelim/2007/r01.htm & r02.htm (preliminary 2007 data)
FRA, Railroad Safety Statistics Annual Report, 1997-2005, Tables 1-1, 5-9.
FRA, Accident/Incident Bulletin, 1980-1996, Tables 19, 36.
Note: Excludes grade crossing accidents.
Track-caused accidents per million train-miles have
dropped 76% since 1980 and 28% since 1990, to a new low.
5
4.84
4
3
1.60
2
1.15
1
0
1980
1985
1990
1995
2000
2007
Sources: FRA website: http://safetydata.fra.dot.gov/Prelim/2007/r01.htm & r02.htm (preliminary 2007 data)
FRA, Railroad Safety Statistics Annual Report, 1997-2005, Tables 1-1, 5-9.
FRA, Accident/Incident Bulletin, 1980-1996, Tables 19, 36.
Note: Excludes grade crossing accidents.
Human factors-caused accidents per million train-miles
have dropped 61% since 1980 and 30% since 1990.
3.5
3.0
3.24
2.5
1.80
2.0
1.5
1.26
1.0
0.5
0.0
1980
1985
1990
1995
2000
2007
Sources: FRA website: http://safetydata.fra.dot.gov/Prelim/2007/r01.htm & r02.htm (preliminary 2007 data)
FRA, Railroad Safety Statistics Annual Report, 1997-2005, Tables 1-1, 5-9.
FRA, Accident/Incident Bulletin, 1980-1996, Tables 19, 36.
Note: Excludes grade crossing accidents.
Brake equipment-related train accidents have
dropped 85% since 1980 and 43% since 1990.
200
187
150
100
49
50
28
0
1980
1985
1990
1995
2000
2005
Sources: FRA, Railroad Safety Statistics Annual Report, 1997-2005, Table 5-9.
FRA Accident/Incident Bulletin, 1980-1996, Table 19.
AAR Analysis of FRA Train Accident Database through 2007.
Note: Includes accidents due to locomotive brake defects.
Brake equipment-related train accident rates have
dropped 86% since 1980 and 56% since 1990.
Accidents per Million Train-Miles
0.3
0.261
0.2
0.1
0.080
0.035
0.0
1980
1985
1990
1995
2000
2005
Sources: FRA, Railroad Safety Statistics Annual Report, 1997-2005, Table 1-1, 5-9.
FRA Accident/Incident Bulletin, 1980-1996, Table 19, 36.
AAR Analysis of FRA Train Accident Database through 2007.
Note: Includes accidents due to locomotive brake defects.
Wheel equipment-related train accident rates have
dropped 85% since 1980 and 54% since 1990.
Accidents per Million Train-Miles
0.5
0.411
0.4
0.3
0.2
0.135
0.1
0.062
0.0
1980
1985
1990
1995
2000
2005
Sources: FRA, Railroad Safety Statistics Annual Report, 1997-2005, Table 1-1, 5-9.
FRA Accident/Incident Bulletin, 1980-1996, Table 19, 36.
AAR Analysis of FRA Train Accident Database through 2007.
Note: Includes accidents due to locomotive wheel defects.
Axle and bearings-related train accident rates have
dropped 78% since 1980 and 54% since 1990.
Accidents per Million Train-Miles
0.3
0.287
0.2
0.138
0.1
0.063
0.0
1980
1985
1990
1995
2000
2005
Sources: FRA, Railroad Safety Statistics Annual Report, 1997-2005, Table 1-1, 5-9.
FRA Accident/Incident Bulletin, 1980-1996, Table 19, 36.
AAR Analysis of FRA Train Accident Database through 2007.
Note: Includes accidents due to locomotive axle or bearing defects.
Truck component-related train accident rates have
dropped 83% since 1980 and 50% since 1990.
Accidents per Million Train-Miles
0.5
0.450
0.4
0.3
0.153
0.2
0.1
0.076
0.0
1980
1985
1990
1995
2000
2005
Sources: FRA, Railroad Safety Statistics Annual Report, 1997-2005, Table 1-1, 5-9.
FRA Accident/Incident Bulletin, 1980-1996, Table 19, 36.
AAR Analysis of FRA Train Accident Database through 2007.
Note: Includes accidents due to locomotive truck component defects.
Grade crossing collisions have declined 74% since
1980 and 52% since 1990.
12,000
10,611
10,000
8,000
6,000
5,715
4,000
2,749
2,000
0
1980
1985
1990
1995
2000
2007
Sources: FRA, Railroad Safety Statistics Annual Report, 1997-2005, Table 1-1.
FRA Highway/Rail Crossing Accident/Incident & Inventory Bulletin, 1980-1996, Table S.
FRA website: http://safetydata.fra.dot.gov/Prelim/2007r01.htm (preliminary 2007 data)
Notes: Includes collisions involving pedestrians. Includes collisions at private crossings.
Grade crossing fatalities have declined 59% since
1980 and 52% since 1990.
1,000
833
800
698
600
400
338
200
0
1980
1985
1990
1995
2000
2007
Sources: FRA, Railroad Safety Statistics Annual Report, 1997-2005, Tables 1-1, 1-3.
FRA Highway/Rail Crossing Accident/Incident & Inventory Bulletin, 1980-1996, Table S.
FRA website: http://safetydata.fra.dot.gov/Prelim/2007/r01.htm (preliminary 2007 data)
Note: Includes pedestrians, employees, passengers, and collisions at private crossings.
Grade crossing injuries have declined
73% since 1980 and 57% since 1990.
4,000
3,890
3,000
2,407
2,000
1,000
1,031
0
1980
1985
1990
1995
2000
2007
Sources: FRA, Railroad Safety Statistics Annual Report, 1997-2005, Tables 1-1, 1-3.
FRA Highway/Rail Crossing Accident/Incident & Inventory Bulletin, 1980-1996, Table S.
FRA website: http://safetydata.fra.dot.gov/Prelim/2007/r01.htm (preliminary 2007 data)
Note: Includes pedestrians, employees, passengers, and collisions at private crossings.
Grade crossing collision rates have declined for 29
consecutive years, 77% since 1980 and 63% since 1990.
Grade Crossing Collisions per Million Train-Miles
16
14.79
12
9.39
8
4
3.47
0
1980
1985
1990
1995
2000
2007
Sources: FRA, Railroad Safety Statistics Annual Report, 1997-2005, Table 1-1.
FRA Highway/Rail Crossing Accident/Incident & Inventory Bulletin, 1980-1996, Table S.
FRA website: http://safetydata.fra.dot.gov/Prelim/2007/r01.htm & r02.htm (preliminary 2007 data)
Note: Includes accidents involving pedestrians and collisions at private crossings.
Since 1980, the total number of public crossings has declined
35%, while the number with gates has increased 146%.
250,000
35% Decrease
200,000
Passive 75%
150,000
Passive 54%
100,000
50,000
Lights 17%
Gates 8%
Lights 17%
Gates 29%
0
1980
2007
Sources: AAR Analysis of December 2007 FRA Grade Crossing Inventory Database.
FRA, Railroad Safety Statistics Annual Report 2005, Table 9-3.
FRA Highway/Rail Crossing Accident/Incident & Inventory Bulletin, 1980, Table 46.
Grade crossing warning device upgrades work. Gates cut the
accident & fatality rates by 93%.
Grade Crossing Collisions per Billion Collision Opportunities
296
300
250
200
No Injury
Injury
Fatal
150
63
100
21
50
0
Passive
Flashing Lights
Gates
Source: AAR Analysis of FRA Grade Crossing Incident & Inventory Databases,
using 1999-2003 incident data and the December 2003 inventory data.
Note: Collision opportunities are measured here as the average number of trains per hour multiplied
by the average number of vehicles per hour moving over each crossing. It is useful primarily
as a measure of relative, not absolute, exposure, since the time period (hour) is arbitrary.
Grade crossing collisions are usually caused by
motorist error.
Stopped,
Proceeded
6%
Stopped on Tracks: 29%
Other
8%
Did Not Stop 41%
Drove Around
Gate: 16%
Sources: AAR Analysis of Highway-Rail Incident Database for 2007 (July 2008).
FRA, Railroad Safety Statistics Annual Report 2005, Table 8-6.
Note: Motor vehicle highway-rail incidents at public crossings.
Trespasser fatalities continue to be a significant
safety problem.
600
543
457
473
450
300
150
0
1980
1985
1990
1995
2000
2007
Sources: FRA, Railroad Safety Statistics Annual Report, 1997-2005, Tables 1-2, 10-3.
FRA Accident/Incident Bulletin, 1980-1996, Table 13.
FRA website: http://safetydata.fra.dot.gov/Prelim/2007/r03.htm (preliminary 2007 data)
Note: Excludes "trespasser" fatalities at grade crossings.
But the trespasser fatality rate per million train miles
was still the lowest in 25 years in 2005.
1.00
0.80
0.89
0.64
0.60
0.60
0.40
0.20
0.00
1980
1985
1990
1995
2000
2007
Sources: FRA, Railroad Safety Statistics Annual Report, 1997-2005, Tables 1-2, 10-3.
FRA Accident/Incident Bulletin, 1980-1996, Table 13.
FRA website: http://safetydata.fra.dot.gov/Prelim/2007/r03.htm (preliminary 2007 data)
Note: Excludes "trespasser" fatalities at grade crossings.
85% of trespasser fatalities in 2007 resulted from
being struck by trains or freight cars.
Train Accidents 3
Other 71
Struck by On-Track
Equipment 410
Source: AAR Analysis of FRA Casualty Database for 2007.
FRA, RR Safety Statistics Annual Report 2005, Table 10-10.
Note: Excludes "trespasser" fatalities at grade crossings.
Many Trespassers are Killed While Intoxicated.
• A North Carolina study for the period 1990-1994 found that
78% of 128 trespassers killed on railroad property were
intoxicated. The median blood alcohol level for this group was
2.5 times the legal limit.
• A 1994 South Carolina study of 24 train-related pedestrian
fatalities found that 79% were intoxicated.
• Toxicology results on 78 of 132 railroad trespassers killed in
Georgia in the period 1990-1996 found 40 (51%) with alcohol
levels above 100 mg/dL.
• A 2005 survey of coroners and Chief Medical Examiners found
that alcohol and/or drugs were judged to be a factor in 530
(57%) of 929 trespasser fatalities in the period 2002-2004 for
which the additional data could be obtained.
Sources: Andrew Pelletier, MD, "Deaths Among Railroad Trespassers: The Role of Alcohol in Fatal Injuries,”
Journal of the American Medical Association, Vol. 277, No. 13, April 2, 1997, pp. 1064-66.
Center for Disease Control, Morbidity & Mortality Weekly Report, Vol. 48, No. 25, July 2, 1999.
Bruce George, “Rail Trespasser Fatalities: Developing Demographic Profiles,” presentation to Transportation
Research Board Annual Meeting, January 15, 2008.
Passengers are 15 to 20 times as likely to die on the
highway as on a train.
Passenger Fatalities per Billion Passenger Miles
16
Highway
12
8
Railroads
Airlines
4
0
1980
1985
1990
1995
2000
2005
Sources: BTS, U.S. Air Carrier Traffic Statistics, Domestic Sched & Unsched; NTSB, Aviation Accident Data Base;
FHWA, Highway Statistics 2006, Table VM-1. NHTSA, Traffic Safety Facts 2006, Early Ed., Table 4, p. 18.
FRA Accident/Incident Bulletin, Tables 13, 36; RR Safety Statistics Annual Report 2005, Tables 1-3, 2-9.
Excludes 232 airline passenger fatalities in the Sept. 11, 2001 terrorist attacks.
Railroads and airlines have comparable passenger
fatality rates.
Passenger Fatalities per Billion Passenger Miles
5
4
3
Railroads
Airlines
2
1
0
1980
1985
1990
1995
2000
2005
Sources: BTS, U.S. Air Carrier Traffic Statistics, Domestic Sched & Unsched; NTSB, Aviation Accident Data Base;
FRA Accident/Incident Bulletin, Tables 13, 36; RR Safety Statistics Annual Report 2005, Tables 1-3, 2-9.
Note: 1993 RR fatalities included 42 killed in one Amtrak accident caused by a barge hitting & misaligning
a RR bridge and another 6 killed by a gunman on the Long Island RR. Excludes 232 fatalities in 9/11 attacks.
Railroads have reduced employee casualty rates by
80% since 1980 and 71% since 1990.
Total Casualties per 100 Full-Time Employees
12
11.16
10
8
7.59
6
4
2.19
2
0
1980
1985
1990
1995
2000
2007
Sources: FRA, Railroad Safety Statistics Annual Report, 1997-2005, Tables 1-2, 4-1.
FRA Accident/Incident Bulletin, 1980-1996, Tables 13, 36.
FRA website: http://safetydata.fra.dot.gov/Prelim/2007/r01.htm & r02.htm (preliminary 2007 data)
Note: Casualties include fatalities as well as injuries and occupational illnesses.
Railroads have lower employee injury rates than do
other major industry groups.
Lost Workday Injuries & Illnesses per 100 Full Time Employees, 2006
4
3
2
1
RA
IL
RO
A
DS
Ag
ric
Co ultu
re
ns
tru
cti
on
Ho
M
an
te
l
uf
ac s
Gr
t
oc urin
er
g
De
y
pa
rtm Stor
es
en
tS
to
re
s
0
Source: Bureau of Labor Statistics, http://www.bls.gov/iif/oshwc/osh/os/ostb1765.pdf
Railroads have lower employee injury rates than do
other transportation modes.
Lost Workday Injuries & Illnesses per 100 Full Time Employees, 2006
8
6
4
2
Tr
an
s
in
e
U
rb
A
irl
it
an
s
ng
rg
e
Ba
uc
ki
Tr
RA
IL
R
O
A
D
S
0
Source: Bureau of Labor Statistics, http://www.bls.gov/iif/oshwc/osh/os/ostb1765.pdf
Railroads have lower employee fatality rates
than do other transportation modes.
Fatalities per 100,000 Full Time Employees, 2006
40
30
20
10
(2
00
5)
W
at
er
Tr
.
uc
ki
ng
Tr
an
sit
Tr
in
es
A
irl
RA
IL
R
O
A
D
S
0
Sources: Bureau of Labor Statistics, National Census of Fatal Occupational Injuries, 2006, Table 2;
And BLS Quarterly Census of Employment & Wages (QCEW) Program for number of employees.
FRA website: http://safetydata.fra.dot.gov/Prelim/2006/r03.htm (preliminary 2006 data).
FRA, Railroad Safety Statistics Annual Report 2005, Tables 1-1 and 1-2.
Railroads have about the same injury severity
as do other industries.
Percent of Lost Workday Employee Injuries, 2006
60
50
40
30
Railroads
Transport
All Industry
20
10
es
s
Fr
ac
Am ture
s
pu
ta
tio
ns
Fa
ta
lit
ie
s
Ill
n
s
O
cc
Cu
t
Sp
ra
in
s
Br
ui
se
s
0
Source: Bureau of Labor Statistics, Table R49. Number of nonfatal occupational injuries and illnesses
involving days away from work by nature of injury and industry division, 2006. http://www.bls.gov/iif/
oshwc/osh/case/ostb1841.pdf . AAR Analysis of FRA Casualty Database for 2006.
See also FRA, Railroad Safety Statistics Annual Report 2005, Table 4-1.
Railroads have about the same injury severity
as do other industries.
Percent of Lost Workday Employee Injuries, 2006 (Sprains not shown)
16
14
12
10
8
6
4
2
0
Fr
ac
tu
re
Am
s
pu
ta
tio
ns
Fa
ta
lit
ie
s
es
s
Ill
n
s
O
cc
Cu
t
Br
ui
se
s
Railroads
Transport
All Industry
Source: Bureau of Labor Statistics, Table R49. Number of nonfatal occupational injuries and illnesses
involving days away from work by nature of injury and industry division, 2006. http://www.bls.gov/iif/
oshwc/osh/case/ostb1841.pdf . AAR Analysis of FRA Casualty Database for 2006.
See also FRA, Railroad Safety Statistics Annual Report 2005, Table 4-1.
U.S. railroads have had lower employee injury rates than
have most major European railroads.
Lost Workday Injuries per 100 Employees
8
Spain
6
Italy
Belgium
4
Austria
France
Norway
BR
Germany
U.S.
Britain
2
Sweden
0
1987
1989
1991
1993
1995
1997
1999
2001
Sources: UIC, Statistics on Accidents at Work, 1993-2001. FRA, Accident/Incident Bulletin, Tables 43, 45, 46.
FRA, Railroad Safety Statistics Annual Reports, 1997-2001, Tables 4-1, 1-1. UIC discontinued these stats after 2001.
Notes: Excludes occupational illnesses. Excludes injuries with no days away from work. Includes fatal injuries.
Data for Sweden & Norway (1993-96) and for U.S. (all years) is per 200,000 hours actually worked, hence higher.
The safest U.S. RRs have had lower employee injury
rates than have the safest European RRs.
Lost Workday Injuries per 100 Employees
4
Norway
3
Sweden
BR
IC
CSX
2
UP
BNSF
NS
1
0
1987
1989
1991
1993
1995
1997
1999
2001
Sources: UIC, Statistics on Accidents at Work, 1993-2001. FRA, Accident/Incident Bulletin, Tables 43, 45, 46.
FRA, Railroad Safety Statistics Annual Reports, 1997-2001, Tables 4-1, 1-1. UIC discontinued these stats after 2001.
Notes: Excludes occupational illnesses. Excludes injuries with no days away from work. Includes fatal injuries.
Data for Sweden & Norway (1993-96) and for U.S. (all years) is per 200,000 hours actually worked, hence higher.
In 2007, train accidents and grade crossing collisions
accounted for less than 4% of employee injuries.
All Other Incidents 96.35%
Train Accidents 1.71%
Grade Crossing Collisions 1.94%
Source: FRA website: http://safetydata.fra.dot.gov/Prelim/2007/r03.htm (preliminary 2007 data)
See also FRA, Railroad Safety Statistics Annual Report 2005, Table 1-3.
Railroads have reduced employee fatalities by 82%
since 1980 and 57% since 1990.
100
97
80
60
40
40
20
17
0
1980
1985
1990
1995
2000
2007
Sources: FRA, Railroad Safety Statistics Annual Report, 1997-2005, Tables 1-2, 1-3, 4-2.
FRA Accident/Incident Bulletin, 1980-1996, Tables 13. Employee on Duty Fatalities.
FRA website: http://safetydata.fra.dot.gov/Prelim/2007/r03.htm (preliminary 2007 data)
Railroads employee fatality rates have fallen 51%
since 1990.
Employee On-Duty Fatalities per 100,000 Employees
20
16
14.5
12
8
7.0
4
0
1980
1985
1990
1995
2000
2007
Sources: FRA, Railroad Safety Statistics Annual Report, 1997-2005, Tables 1-1, 4-2.
FRA Accident/Incident Bulletin, 1980-1996, Tables 13, 36.
FRA website: http://safetydata.fra.dot.gov/Prelim/2007/r03.htm & r02.htm (preliminary 2007 data)
Of 21 employee fatalities in 2007, 12 were due to being
struck or crushed by moving on-track equipment.
Transp,
Struck
by MOW Struck by Train - 5
Switching, Struck Train
Or Crushed By Cars or
-1
Locos. (“SOFA”)- 6
Road TrainGrade
Collision Xing
- 2 Collision
- 1
MOW – Other
-4
MOE – Electrocution - 1
MOE – Highway Accident - 1
Sources: News clippings and FRA, Monthly Reports on Employee Fatalities, 2007.
FRA, Railroad Safety Statistics Annual Report, 1997-2005, Tables 1-2.
Of 16 employee fatalities in 2006, none occurred in train
accidents, 4 occurred in grade crossing collisions, and 9
were due to being struck by moving on-track equipment.
Transp
Switching, Struck
Or Crushed By Cars or
Locos. (“SOFA”)- 5
MOW Struck
by MOW Equip
MOW In Hwy
-2
Accident - 1
Other – 1
Veh struck bldg.
Transp in Grade
MOE
MOE
Crossing Collision
MOE in Struck or
in Veh
-2
Grade Xing Crushed by Frt Collision - 1
Collision - 2 Cars - 2
Sources: News clippings and FRA, Monthly Reports on Employee Fatalities, 2006.
FRA, Railroad Safety Statistics Annual Report, 1997-2005, Tables 1-2.
Of 25 employee fatalities in 2005, 11 were due to being
struck or crushed by moving on-track equipment and 8 were
in train or grade crossing collisions.
Switching, Struck
Or Crushed By Cars or
Locos. (“SOFA”)- 11
Road Train
Collisions
-7
Transportation:
Train Accident
Transportation,
Not in Train
Accident
MOW Struck by Train - 2
Grade
Xing
Collision
- 1
MOW
In Hwy
Accident
-2
Maint. of Way
Explosion - 1
Electrocution - 1
Maint of Equip.
Sources: News clippings and FRA, Monthly Reports on Employee Fatalities, 2005.
FRA, Railroad Safety Statistics Annual Report, 1997-2005, Tables 1-2.
Hazmat accident rates have declined 87% since 1980
and 34% since 1990.
Train Accidents with a Release per Thousand Hazmat Carloads
0.16
0.14
0.143
0.12
99.996% of Carloads
are Accident Release Free
0.10
0.08
0.06
0.04
0.02
0.018
0.027
0.00
1980
1985
1990
1995
2000
2005
Sources: FRA, RR Safety Statistics Annual Report, 1997-2005, Table 6-1. FRA, Accident/Incident Bulletin,
Table 26. AAR Analysis of 2006 FRA Train Accident Database. Carloads from ICC/STB Waybill Sample, 1995-2006.
Terminated carloads adjusted to counter known hazmat underreporting. Prior to 2003 terminated carloads overestimated
by 10% to 15% due to inability to exclude duplicates associated with rebilling.
Accidents with a hazmat release have
declined 61% since 1980.
Reportable Train Accidents with a Hazmat Release
140
120
119
100
80
60
46
40
35
20
0
1980
1985
1990
1995
2000
2007
Sources: FRA, RR Safety Statistics Annual Report, 1997-2005, Tables 6-1. FRA, Accident/Incident Bulletin,
Table 26. AAR Analysis of 2007 FRA Train Accident Database.
Note: An accident may involve releases from more than one car.
Hazmat incident release rates have declined 72%
since 1980 and 58% since 1990.
Incidents per Thousand Hazmat Carloads
1.6
1.52
1.4
1.2
0.99
1.0
0.8
0.6
0.42
0.4
99.96% Incident Free
0.2
0.0
1980
1985
1990
1995
2000
2005
Sources: USDOT, Pipeline & Hazardous Materials Safety Administration, Hazardous Materials Incidents
by Year & Mode. http://hazmat.dot.gov/files/hazmat/10year/10yearfrm.htm for 1997-2006.
Includes releases in train accidents as well as non-accident releases. ICC/STB Waybill Sample.
1995-2006. Terminated carloads adjusted to counter known hazmat underreporting. Prior to 2003 terminated
carloads overestimated by 10% to 15% due to inability to exclude duplicates associated with rebilling.
Over the last 10 years, about 1.1% of train accidents have
resulted in a release of hazardous materials.
No Hazmat Cars Present
23,715 75.5%
Hazmat Cars in
Consist 3,997 12.7%
Hazmat Cars
Derailed or
Damaged 3,360
10.7%
Hazmat Released 354 1.1%
Source: AAR Analysis of 1998-2007 FRA Train Accident Databases, July 2008.
Note: Includes grade crossing train accidents meeting dollar threshold.
In 2007, about 1.6% of reportable train accidents
resulted in a release of hazardous materials.
No Hazmat Cars Present
2,098 73.3%
Hazmat Cars in
Consist 416 14.5%
Hazmat Cars
Derailed or
Damaged 302
10.6%
Hazmat Released 46 1.6%
Source: AAR Analysis of Year 2007 FRA Train Accident Database, July 2008.
Note: Includes grade crossing train accidents meeting dollar threshold.
Rail freight transport incurs about 12% of the
fatalities that trucks do per trillion ton-miles.
4,000
3,761
3,207
3,500
3,000
2,500
Fatalities
Fatality Rate
2,000
1,500
720
391
1,000
500
0
Rail Freight
Tractor-Trailers
Sources: AAR Analysis of FRA Casualty Database for 2006. Rail Ton-Miles in 2006 from
RR Facts, 2007, pp. 27, 32. USDOT, FMCSA, Large Truck Crash Facts, 2006, Table 13,
http://ai.volpe.dot.gov/CarrierResearchResults/PDFs/LargeTruckCrashFacts2006.pdf (p. 19)
Tractor-trailer net ton-miles in 2002 estimated at 1.14 trillion from 2002 USDOC,Vehicle Inventory
& Use Survey (VIUS), updated to 2006 based on VMT from same Table 13.
Rail freight transport incurs about 6% of the
injuries that trucks do per trillion ton-miles.
60,000
56,000
48,000
50,000
40,000
Injuries
Injury Rate
30,000
20,000
5,419
2,942
10,000
0
Rail Freight
Tractor-Trailers
Sources: AAR Analysis of FRA Casualty Database for 2006. Rail Ton-Miles in 2006 from
RR Facts, 2007., p. 27, 32. USDOT, FMCSA, Large Truck Crash Facts, 2006, Table 15,
http://ai.volpe.dot.gov/CarrierResearchResults/PDFs/LargeTruckCrashFacts2006.pdf (p. 22)
Tractor-trailer net ton-miles in 2002 estimated at 1.14 trillion from 2002 USDOC,Vehicle Inventory
& Use Survey (VIUS), updated to 2006 based on VMT from Large Truck Crash Facts Table 13.
Railroads incurred 17 fatalities in the last 10 years
due to hazmat while trucks incurred 124.
140
120
100
80
60
40
20
0
Rail
Truck
Source: USDOT, Pipeline & Hazardous Materials Safety Administration, Hazardous Materials Incidents
by Year & Mode, from http://hazmat.dot.gov/pubs/inc/data/tenyr.pdf for 1998 through 2007,
as of 6/11/2008.
Railroads now have less than 5% of the hazmat incidents
that trucks have, despite roughly equal hazmat ton-mileage.
18,000
16,000
14,000
12,000
10,000
8,000
Truck
Railroad
6,000
4,000
2,000
0
1988 1990 1992 1994 1996 1998 2000 2002 2004 2006
Sources: USDOT, Pipeline & Hazardous Materials Safety Administration, Hazardous Materials Incidents
By Year & Mode, from http://hazmat.dot.gov/pubs/inc/data/10yearfrm.htm for 1998 through 2007. 6/11/2008
In 2005, the truck and rail modes hauled about the same number of hazmat net ton-miles, estimated at 114 billion
for trucks and 111 billion for rail. USDOC, 2002 Commodity Flow Survey (CFS), Table 1a, for truck hazmat
ton-miles, updated using combination truck VMT from FHWA Highway Statistics,Table VM-1.
STB Waybill Sample for rail ton-miles.
Railroads now have 21% of the serious hazmat incidents that
trucks have, despite roughly equal hazmat ton-mileage.
500
450
400
350
300
250
200
150
100
50
0
1989 1991 1993 1995 1997 1999 2001 2003 2005 2007
(New definition
of “Serious”)
Truck
Railroad
Sources: USDOT, Pipeline & Hazardous Materials Safety Administration, Hazardous Materials Incidents
by Year & Mode, from http://hazmat.dot.gov/pubs/inc/data/tenyr_new_serious.pdf, as of 6/11/2008.
PHMSA’s new definition defines serious incidents as those involving a fatality or serious injury due to
a hazmat release or evacuation of 25 or more people as the result of a hazmat release or fire.
In 2005, the truck and rail modes hauled about the same number of hazmat net ton-miles, estimated at 114 billion
for trucks and 111 billion for rail. USDOC, 2002 Commodity Flow Survey (CFS), Table 1a, for truck hazmat
ton-miles, updated using combination truck VMT from FHWA Highway Statistics,Table VM-1. STB Waybill Sample.
Railroads now have less than 18% of the hazmat accidents
that trucks have, despite roughly equal hazmat ton-mileage.
400
350
300
250
200
Truck
Railroad
150
100
50
0
1994
1996
1998
2000
2002
2004
2006
Sources: USDOT, Pipeline & Hazardous Materials Safety Administration, Hazardous Materials
Incidents by Year & Mode, from http://hazmat.dot.gov/pubs/inc/data/tenyr_accd.pdf. 6/11/2008
In 2005, the truck and rail modes hauled about the same number of hazmat net ton-miles, estimated at 114 billion
for trucks and 111 billion for rail. USDOC, 2002 Commodity Flow Survey (CFS), Table 1a, for truck hazmat
ton-miles, updated using combination truck VMT from FHWA Highway Statistics,Table VM-1.
STB Waybill Sample for rail ton-miles.
Technological Improvements to Railroad Safety:
Track & Equipment
Improved Track
• Rail steels
• Welded rail
• Fasteners
• Detection of flaws, weak spots
Improved Equipment
• Heat treated curved plate wheels
• Hot box detectors, roller bearings, acoustic detection
systems
• Air brake control valves & air brake tests
• Advanced Technology Safety Initiative (ATSI), to
identify and repair high impact wheels & other flaws.
Over 3.6 years, ATSI has prevented an estimated 160
broken rail and broken wheel accidents on main track.
0.4
Main Track Accidents per Million Freight Train Miles
Without ATSI (Estimate)
0.3
With ATSI (Actual)
0.2
0.1
“Before” Period
(Pre-ATSI)
“After” Period
(ATSI)
0.0
1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007
Sources: AAR Analysis of FRA Train Accident and Train-Mile Data, 1997-2007,
All U.S. Railroads.
Technological Improvements to Railroad Safety:
Tank Cars and SNF
•
•
•
•
Head shields and shelf couplers
Thermal insulation
Bottom outlet protection
Enhanced safety requirements of hazmat tank cars >263,000
lbs or carrying environmentally sensitive chemicals
• Enhanced standard for cars carrying spent nuclear fuel and
high level radioactive waste.
• A new performance standard for TIH tank cars.
Railroad Industry Safety Programs:
Hazardous Materials
• AAR North American Non-Accident Release (NAR)
Program
• Transportation & Community Awareness & Emergency
Response (TRANSCAER)
• ACC Responsible Care
• Operation Respond
• TTCI's Emergency Response Training Center
• TTCI's BOE Hazmat Inspections
• TTCI's BOE Hazmat Safety Information
Railroad Industry Safety Programs:
Operations, Training, Crossings
•
•
•
•
•
•
•
•
•
Individual Railroad Employee Safety Programs
Crew Resource Management (CRM)
Peer Observation Programs
Fatigue Countermeasures
Remote Control Operations
Locomotive Simulators
Interactive Video Individual Training
Operation Lifesaver
Grade Crossing Upgrade (Section 130) Program
Railroad Security
• After the 9/11 terrorist attacks, the AAR & its member
railroads voluntarily conducted a comprehensive risk
assessment and implemented a security action plan with
four alert levels.
• The Security Plan encompasses over 1,300 critical
facilities (tunnels, bridges, yards, dispatch centers, etc.)
over a 142,000-mile nationwide rail network.
• Execution of the Security Plan depends on timely receipt
of threat & warning information from the federal govt.
• Former DHS Secretary Ridge cited this plan as a model
for other U.S. industries.
Railroad Security:
Countermeasures include:
• Employee and customer awareness and training: e.g. to thwart
terrorist intelligence gathering, facilitate warning & recovery.
• Sharing of Intelligence on Threats:
– AAR Operations Center (24/7) collects, evaluates, and disseminates
information to railroads thru the Railway Alert Network (RAN)
– The Surface Transportation Information Sharing & Analysis Center (STISAC) collects, analyzes, and disseminates information on physical and
cyber-security threats to providers and users of surface transportation.
•
•
•
•
Vetting employees & contractors.
Controlling access to critical facilities and shipment info.
Special operating measures to safeguard hazmat.
Securing communications and data and ensuring message
integrity and best IT practices.
Since 1980, U.S. Class I railroads have spent $396
billion on track and equipment.
Capital Expenditures and Maintenance Expenses, Minus Depreciation, in Billions of Dollars
12.0
10.0
Equipment
8.0
6.0
Track & Structures
4.0
2.0
0.0
1980
1985
1990
1995
2000
2007
Sources: AAR, Analysis of Class I Railroads, 1980 - 2007, based on R-1 Reports submitted by each Class I
railroad to the ICC/STB. Equipment: Lines 382+158-151-154-157. Track: Lines 378+149-147.
Note: Current year dollars.
Since 1991, U.S. Class I railroads have accelerated
capital spending on new rail.
Billions of Current Dollars
2.5
2.0
Capital Expenditures on Rail
1.5
1.0
0.5
Operating Expenses for Rail & Other Track Material
0.0
1987 1989 1991 1993 1995 1997 1999 2001 2003 2005 2007
Source: R-1 Reports submitted by each Class I railroad to the ICC/STB, 1987-2007.
Schedule 330, line 8, column e; Schedule 410, lines 1, 14, and 15, column h.
In the last 10 years, U.S. Class I railroads have laid
5.9 million tons of new rail.
0.75 Million Tons of New Rail Laid
0.50
0.25
0.00
1987 1989 1991 1993 1995 1997 1999 2001 2003 2005 2007
Note: Includes new rail laid in replacement and in addition, excludes relay rail.
Sources: AAR, Analysis of Class I Railroads, 1980 – 2007, Lines 366+369+371,
based on R-1 Reports submitted by each Class I railroad to the ICC/STB.