D) Other High Risk Construction Activities (but not limited to): o Personnel entering trenches more than 1.5 metres in depth. O Use of explosives (excluding powder-actuated hand-held fastening tools). O Work near an exposed energised electrical installation. O Work on or adjacent to a road. O Using confined spaces. O Using hazardous substances. Sample Needs Assessment - 9+Free Word, PDF Documents. 10+ Audit Memo Templates – Free Sample, Example, Format. 15+ Property Inventory Templates - Free Word, PDF, Excel. Business Continuity Plan - 9+ Free PDF, Word Download Document. Security Policy Template - 7 Free Word, PDF Document Downloads. Advised that I may request that bed rails be installed on the resident’s bed. The risk and alternatives to using bed rails, as they apply to this resident’s particular condition and circumstances, have been clearly explained to me. I understand that, in addition to this signed consent form authorizing the use of bed rails for this.
1d41fbd7b384b93.pdf, 6.pdf, eleanor oliphant is completely fine free pdf, royal high school edinburgh james hutton, electricity freedom system pdf free download, portland cement plaster (stucco) manual, d and d spell casting ability, minecraft guide book mod, 4.pdf, towujapiruni.pdf, happy.
Monorail System PPT, PDF Report Free Download: There are two Types of Monorail System and they are Straddle type and Suspended type.
- Magnetic Levitation (Maglev)
- Requires electromagnetic current to generate magnetic field.
- Superconducting electric magnets in the vehicle to drift and move the train
- An AC is ran through electromagnet coil
- Creates a magnetic field that enchant and repels the superconducting magnets
- Revesring the direction of AC: braking.
- Varying intensity of AC : speed change
Also See: Bio-Battery Seminar and PPT with PDF Report
Monorail System PPT, PDF Report Free Download
- Dig a hole
- Drop in a pre-built support pylon,
- Truck in the track which was manufactured offsite,
- Lift into place.
- Monorail can be built faster with less cost.
- Consumes minimal space.
- Ideal mode of transportation for population less then 3 lakh
- Serves as Feeder lines for denser cities
- Environmental friendly exert no carbon derived fuels
- No risk of vehicles getting struck in crossings
- Practically silent while travelling
- Contributes Greener Environment
- Regularly operates at 99.99% reliability.
- To replace a section of track, the entire system needs to be shut down.
- The capacity-to-cost ratio is less, even though the per kilometer erection cost is less.
- Cannot run without electricity.
- Monorail tracks do not easily accommodate at-grade intersections.
- In an emergency, instant exit is not possible
- The biggest disadvantage cited in the case of monorail is its limited passenger capacity.
Also See: Black Box Testing Seminar and PPT with PDF Report
Content of the Seminar and pdf report for Monorail System
- HISTORY OF MONO RAIL
- WORKING PRINCIPLE
- TYPES OF TRANSIT SYSTEM
Here we are giving you Monorail System Seminar and PPT with PDF report. All you need to do is just click on the download link and get it.
It was all about Shape Monorail System Seminar and PPT with pdf report. If you liked it then please share it or if you want to ask anything then please hit comment button.
Risky Rails Pdf Free Download For Windows 7https://studymafia.org/monorail-ppt-pdf-report-free-download/ECE SeminarsMonorail System PPT, PDF Report Free Download: There are two Types of Monorail System and they are Straddle type and Suspended type.Working PrincipleMagnetic Levitation (Maglev)Requires electromagnetic current to generate magnetic field.Superconducting electric magnets in the vehicle to drift and move the trainAn AC is ran...Sumit ThakurSumitThakur[email protected]AdministratorI am an Indian Blogger. I am passionate about blogging. If you want to ask me anything about blogging then feel free to ask 🙂Study Mafia: Latest Seminars Topics PPT with PDF Report 2021
Risky Rails Pdf Free Download 64 Bit
Railroad Injury: Causes, Costs, and Comparisons with Other Transport Modes Ted R. Miller, John B. Douglass,
and Nancy M. Pindus
Safety regulators in the railroad industry need comprehensive injury costs by cause, and cost comparisons to other modes of transport to gain policy insights. This study links case-level data from the Federal Railroad Administration to monetary costs and consequences of specific injuries. Lost quality of life was estimated from physicians’ ratings of injury impairment and the value of fatal risk reduction. Railroad injuries and illnesses cause an average of $3 billion in fatality costs and $650 million in nonfatal incident costs annually. Impacts at rail-highway crossings impose three times as much cost as the next highest cause. Commercial air travel is the safest per passenger mile. Trains, second lowest, have injury costs six times higher, but only one-fifth as high as cars. The relative safety of trains and airplanes in transport means highway modes enjoy implicit price subsidies. Better injury cost recovery through taxation or insurance would make the subsidies explicit. More informed modal choices and thus enhanced economic efficiency would result.
INTRODUCTION Ted Miller directs the Safety and Health Policy Program and the Children’s Safety Network Economics and Insurance Resource Center at the National Public Services Research Institute in Landover, MD. Previously, he directed the Safety and Risk Policy Program at the Urban Institute for many years. A safety economist, Miller received his doctorate from the University of Pennsylvania in 1975. Miller has published extensively on injury costs. He is a member of the editorial boards of the Journal of Safety Research, Accident Analysis and Prevention, and the Journal of Forensic Economics. His crash cost estimates are used by Transport Departments in the U.S., Canada, and New Zealand. Upon receiving his Masters in Economics from the University of Wisconsin in 1989, John Douglass joined the Urban Institute’s safety policy group. Douglass became an independent consultant in 1993. Nancy Pindus received a MBA from the Leonard Davis Institute of Health Economics at the University of Pennsylvania in 1973. Since then, she primarily has worked as a health policy analyst and program evaluator. Ms. Pindus joined the Urban Institute as a Senior Research Associate in 1989.
Winter 1994Nolume 25/Number 4
Little is known about the full extent of injury costs and consequences in the railroad industry. French (1990, 1992) estimated railroad workers’ willingness to pay to avoid nonfatal injury using a wage-risk regression model. But since different accidents and injuries may have very different human consequences, more detailed cost estimates are needed to help regulators allocate rail safety resources rationally and perform benefit-cost analyses of proposed safety interventions. This paper expands our understanding of costs for railroad injuries in several ways. First, we investigate costs for all railroad injury and illness including fatalities. This aggregate includes not only worker injuries, but injuries to passengers, contractors, other 183
nontrespassers, and trespassers. Only incidents involving operation or servicing of a train or maintenance of way, structures, etc. are considered. Second, in summary tables, we provide the various cost components: medical (including ancillary}, injury claims administration, wage and household production loss, lost quality of life, and the sum of these, the comprehensive injury cost. Third, we explore cost breakdowns, including crash costs by type of train and type of incident and total injury costs by cause. Finally, we compare injury costs by mode of transportation. Costs per passenger mile and per vehicle mile are estimated for railroad, truck, bus, car, commercial air, and general aviation. We also provide cost estimates per ton mile for truck versus railroad. Railroad costs were estimated by linking case-level data from the Federal Railroad Administration (FRA) to several public data sets describing the costs and consequences of injuries by injury. This allowed computation of estimates crossbred from the various data sets. METHODS
This section focuses on the methods used to determine the cost of railroad injuries. Costs for other modes came from our studies (Miller, Viner, & Associates, 1991; Miller, 1993). The costing methods were essentially the same, except different data sources were tapped to learn injury incidence, hospitalization rates, lengths of hospital stay and days of work lost. Data Sources Injury data are not collected uniformly. Almost every national data collection agency codes injury descriptions differently. Our data file was built from several public use files, and this required matching codes and merging data across several systems. Data sources used to develop the railroad estimates are described below. The specific data used to develop cost estimates are discussed in the following section. The FRA Accident/Incident Reporting System. FRA’s Office of Safety maintains a 184
database obtained through monthly accident/ incident reports submitted by the railroads. The data include information about all fatal injuries; injuries to persons other than railroad employees if they require medical treatment; injuries to employees if they require medical treatment and result in lost workdays or restriction of work, transfer or termination, or loss of consciousness; and all occupational illnesses diagnosed by physicians. The information reported includes a description of the injury or occupational illness, the circumstances of the injury, and for employees, the number of days away from the job or on restricted duty. FRA injury codes include a 2digit body part and a 2-digit nature of injury. DCI ~Detailed claims I?~f~~ati~n~. This is a proprietary database of the National Council on Compensation Insurance. This longitudinal file is a nationally representative sample of Workers Compensation lost workday claims spanning 17 states from 1979 to 1988. It contains data on 452,000 injuries, 138,000 of them hospitalized. Insurers report on claims in the DC1 sample six months after the injury and annually thereafter until the claim is closed (meaning no more charges are anticipated or a reserve was set aside - and reported to DC1 - to cover predictable future costs). DC1 claims are reopened if unanticipated payments arise after closure. The DC1 record includes hospital payments (no copay is required in the system); medical payments including prescriptions, equipment, and long-term care; vocational rehabilitation payments; length of stay if hospitalized; disability compensation; and classification as total temporary, partial permanent, or total permanent disability. NEISS (the emotional Electronic Injury S~~eillan~e System). This system is maintained by the Consumer Product Safety Commission (CPSC). It gathers nationally representative counts of hospital emergency room visits and admissions for selected causes. The file includes data from less than 100 hospitals. Although NEISS annually collects data on about 200,000 injuries related to consumer products, our comparisons with other files show the sample is too small to yield reliable data for hospitalized injuries. We used Journal of Safety Research
NEISS incidence data on consumer product injuries in 1987-1989 and workplace injuries in 1983-1985. NHIS (the National Health Interview Survey). This file contains nationally representative household interview data on injuries including 3-digit International Classification of Diseases (ICD) code, medical treatment, and resulting productivity losses. We used a 1984-1986 NHIS data set developed by Rice, MacKenzie, and Associates (1989). NHIS ICD codes come from peoples’ imperfect descriptions of their injuries. CHAMPUS (the Civilian Health and Medical Program of the Uniformed Services reimbursement file). This file contains data on hospital and medical utilization, payments, and self-pay for roughly 2,000,OOO military dependents and retirees. CHAMPUS produces annual hard copy summary reports that provide average cost and utilization data for inpatient and outpatient care by 3-digit ICD code. Hospital services and professional fee reimbursements are shown separately on the reports. CHAMPUS covers few males aged 18-45 and few people over age 65. We averaged CHAMPUS data for 1986-1989, with payments inflated to 1989 dollars. NASS (the National Accident Sampling System). NASS is maintained by the National Highway Traffic Safety Administration. It provides a 60-day follow-up on injuries in a sample of police-reported crashes. The sample was most reliable from 1982-1985. We created a summary file for these years. The file includes number of cases and days of work lost per worker by hospitalization status and length of stay if hospitalized. We supplemented the file with physician ratings of the functional capacity typically lost to each injury (Miller et al., 1991). In earlier work, we developed equivalency tables between groups of NASS injury codes, DC1 injury codes, ICD codes, and NEISS injury codes. Then we converted summary data from all the files into data by NEISS codes that were linked to FRA codes (Miller, Pindus, Douglass, & Rossman, 1994). Other data from DC1 were mapped directly to FRA. The NEISS, DC1 and FRA coding systems are very similar. Winter 1994Nolume 25/Number 4
Formulas, Assumptions, and Documentation The costs resulting from railroad injuries cover: Medical and ancillary care Ambulance services Vocational rehabilitation Lost quality of life Lost wages and fringe benefits Lost household production Administrative costs including legal defense costs related to liability insurance claims We call the sum of these costs comprehensive injury costs or societal willingness to pay. The lost quality of life, lost wages and fringe benefits, and lost household production together equal individual willingness to pay. Our data on injury costs and consequences were developed separately for hospitalized and nonhospitalized cases. Unfortunately, FRA does not collect hospitalization status. To combine the data, we weighted them by the fractions hospitalized and not hospitalized by each body part and injury combination. To estimate the fractions, we used combined counts by body part, injury, and hospitalization status from NEISS worker injury data (which excludes injuries in highway crashes), NEISS consumer product injury data, and NASS data on injury in highway crashes. Together, we estimate these files cover 75 to 85% of nonfatal unintentional injuries treated in hospitals and emergency rooms. We could not simply use NEISS worker injury data because the sample included too few hospitalized cases to yield reliable fractions hospitalized by NEISS injury code. Similarly, we could not use the fractions from DC1 because the data are restricted to injuries resulting in several days of work loss. Finally, we chose not to use the NHIS fractions because the sample is too small to yield reliable hospitalized counts by injury code. The NEISYNASS fraction hospitalized was unavailable for occupational illnesses, cumulative injuries, and a few other injuries. In these cases, we used the DC1 fraction hospitalized for the IX4 code times the ratio of the fraction hospitalized across all injuries in the NEISS/NASS 185
file divided by the fraction hospitalized across all illnesses and injuries in the DC1 tile. Medical and ancillary care. The medical payments per death came from Rice et al. (1989). For hospitalized cases, medical “costs” equal medical, hospital, and ancillary payments from the DCI. The DC1 data represent expected lifetime payments by the Workers’ Compensation carrier. Because the inflation rate for medical costs averages roughly 4% above the general inflation rate, we assumed excess inflation exactly offset the 4% discount rate used to compute the present value of DC1 medical payments for years subsequent to the year of the injury. For nonhospitalized cases, medical “costs” were estimated in two steps. The estimates combine data from CHAMPUS, NHIS, DCI, NEISS, and the National Medical Care Utilization and Expenditure Survey (NMCUES). The first step was to compute the CHAMPUSbased lifetime medical payments for the 3digit ICD code encompassing the injury of interest as: CHAMPUS Med =
Both CHAMPUS and NHIS data are at the 3-digit ICD code, sometimes coarser than the codes used by NEISS and FRA. The second step broke the cost estimates down to NEISS codes by multiplying them times the ratio of DC1 medical cost for the NEISS code divided by DC1 cost for the ICD code. To compute the average DC1 medical payments per case for the ICD code and the NEISS code, we weighted the payments per case for each injury in the code group by the number of DC1 cases. Some FRA codes did not match the NEISS codes. In these cases, we used the DC1 nonhospitalized medical payments for the FRA code times the ratio of the average nonhospitalized medical payments across all injuries in the NEISS file divided by the average across all illnesses and injuries in the DC1 file. Ambulance and vocational rehabilitation costs. Probabilities of ambulance transport and costs per transport by hospitalization status and survival came from Rice et al. (1989). We did not cost police or fire services. Vocational rehabilitation costs per medically treated case came from the DC1 data.
FracMedTreat * (STMed + Ancill) * (1 + Frac6Mos+)
where, FracMedTreat = the fraction medically treated from NHIS STMed
= the medical payments in the calendar year of the injury (on average, six months of payments) from CHAMPUS
= ancillary payments from NMCUES. Following Rice et al. (1989), we assumed all cases would have prescription payments. We used the NMCUES ratio of cases with other ancillary payments to prescription payments. NMCUES data were broken down into 13 nature of injury categories, but did not vary by body part injured.
= the fraction of medical payments beyond six months from DCI.
Quality of life and productivity losses. Willingness to pay estimates the value of reducing death and injury risk. It includes the value of reducing productivity losses, pain, suffering, and lost quality of life. Willingness to pay is computed from safety behavior. Although the economics literature named this method “willingness to pay,” that name is misleading. Most empirical estimates in the literature are based on amounts people actually pay to reduce safety risks by small amounts (e.g., 1 in 10,000) not what they are willing to pay (Miller 1990). The U.S. Office of Management and Budget (1989) prescribes using a willingness to pay method in regulatory benefit-cost analyses that place a dollar value on saving human life. Values that people place on slightly reducing fatal risks come from studying how much people pay for small changes in their survival probabilities. The studies fall into four classes: . Wage-risk studies, which analyze compen-
sating wage differentials risky jobs.
Journal of Safety Research
Market studies, which analyze the market for products that affect health and safety. For example, Miller (1990) estimates the value of risk reduction from the rising sales figures for smoke detectors as their price dropped over time. Behavioral studies, which examine riskavoidance behavior in inherently risky situations. For example, Melinek (1974) studied the value implied by a choice between using a pedestrian underpass or saving time by walking through a traffic circle. Surveys, which probe how much people are willing to pay for small changes in risk.
Normally, the value of fatal risk reduction is converted to a value per statistical life. For example, suppose a study estimates that the average person spends $220 on optional auto safety features that reduce their perceived chance of dying prematurely by 1 in 10,000. Dividing $220 by the 1 in 10,000 probability yields a $2.2 million value per statistical life. Miller (1990) critically reviewed the literature. He found a mean value from 47 reasonably sound studies of $2.2 million per statistical life saved (in 1988 dollars), with a standard deviation of $.6 million. Miller adjusted the values to reflect the risk reduction that people perceived they were buying, not actual risk reduction. Where appropriate, the studies also account for the effects of legal mandates on usage levels. To value the quality of life associated with nonfatal risk reduction, we multiplied the value of fatal risk reduction times the ratio of the years of functional capacity at risk in a fatality versus the injury of interest. Then we subtracted the monetary component of this value - the after-tax earnings and household production. The theoretical justification for computing willingness-to-pay to prevent a statistical injury as the product of willingness-to-pay per statistical life times the percentage of functional capacity lost to the injury is provided in Miller, Calhoun, and Arthur (1989). This method requires calibrating functional capacity loss so that the utility of a year of life and a year of functioning are equal. This method assumes that having a capacity loss for 20 years is 20 times as bad as having it for one year. That assumption, although the best possible given Winter 1994Nolume 25/Number 4
the paucity of information on the subject, is questionable (Sackett & Torrance, 1978). Functional capacity loss was defined as impairment along any of seven dimensions: (1) mobility, (2) cognitive, (3) self care, (4) sensory, (5) cosmetic, (6) pain, or (7) ability to perform household responsibilities and wage work. Loss on the first six dimensions was rated by physician experts; the seventh came from the DCI. More specifically, to estimate the years at risk to different injuries, we: l
Used physician estimates in Hirsch et al. (1983) of average capacity lost over time, by injury, along the first six dimensions of functioning. The major drawback of these ratings is that each injury was rated by only one physician. Used a literature review to develop weights and applied them to combine the ratings on seven dimensions into a single rating of the percentage of utility lost over time (see Miller et al., 1989; Miller, 1993; Miller et al., 1994). The weights both convert from functional capacity losses to utility losses, and reflect the relative importance of different aspects of functioning. Added ratings for injuries not rated in Hirsch et al. In some cases, we assumed the functional capacity lost equalled the rated loss for similar injuries; for example, we set the loss for “knee fracture and dislocation” equal to the larger of the losses for “knee fracture” and “knee dislocation.” We assumed the percentage lost in the first year equalled the percentage of work lost during that year for minor (AIS 1) injuries and more serious lacerations. If a person with minor injuries permanently lost earning power, they obviously also lost some capacity to function physically. We assumed, conservatively, that they lost 5%. That equates to mild sensory or cognitive loss; mild losses in mobility or self-care involve a larger percentage loss. Computed the years at risk to the injury from the percentage lost over time and the expected life span.
For occupational illness, we computed only the quality of life loss resulting from workrelated disability. We lacked physician ratings of the capacity lost along other dimensions. 187
Lost earnings including fringe benefits. Lost earnings include both short-term losses during the acute recovery from injury and long-term losses due to consequent permanent disabilities. The short-term losses were computed from the FRA data on workdays lost. The short term loss for railroad workers and contractors equals the average days lost by railroad workers and contractors with this injury, as reported to FRA, times the daily average wage (excluding overtime) for railroad workers. The average daily wage was computed using FRA data on the occupational mix of injured workers and wage data from the Association of American Railroads $114.26. The percentage added for fringe benefits was 37.13% obtained from the Association of American Railroads. For others, we assume the short term loss equals the average days lost for railroad workers times the fraction of U.S. noninstitutionalized residents employed - .626 (Bureau of Labor Statistics, 1989) times the average daily wage for nonsupervisory workers - $17.28 (Economic Report of the President, 1991). We added fringe benefits of 20% (Economic Report of the President, 1991). The long term wage loss equals: Lifetime Earnings * (1 + fringe benefits rate) * [p(Tot Perm Disability) + .17 * p(Part Perm Disability)]. Lifetime earnings were computed from the age distribution of the injured. We assumed 76% of injured railroad workers and contractors are men, the same percentage as in the
participation. We did not use railroad earnings for railroad workers because these workers would not necessarily continue to work in the railroad industry. As a compromise, we did use the much higher railroad fringe rate rather than the general fringe rate for the railroad workers. The probabilities of total permanent and partial permanent disability were computed from the DC1 data. Since virtually any hospitalized injury causes several days of work loss, we took the probabilities for hospitalized injury directly from DCI. We also took the probabilities of permanent disability for nonhospitalized cumulative injuries and occupational illnesses directly from DCI. For nonhospitalized acute injury, we needed to know what percentage of cases would involve enough work loss to qualify for workers’ compensation. We assumed that any injury involving that much work loss would be seen in a hospital emergency room. Especially for back and neck injuries, this assumption may lead to an overestimate of disability probabilities. Nevertheless, it makes it possible to compute probabilities from 1983-1986 NEISS workplace injury incidence data. We developed the following formula for estimating the probability of disability of nonhospitalized workers: P (disability for NEISS Code Group) = Nonhosp Disabled Workers for DC1 Code .____ Nonhosp Workers in NEISS Code Group where,
Nonhosp Workers in NEISS Code Group = Max (A,B) A=
Hosp Workers for DC1 Code - Hosp Workers for DC1 Code Fraction Workers Hosp in NEISS Code Group
B = Nonhospitalized Workers for DC1 code. DC1 data on worker injury. We assumed half of other injured people are men. We computed lifetime earnings at a 4% discount rate from the data in Rice et al. (1989). These data account for likely labor force participation and the probability of employment given 188
Data indicating the loss in earnings for partial permanent disability averages 17% come from Berkowitz and Burton (1987). Lost household production. We used values for household production from Douglass, Journal of Safety Research
Kenney, ed that:
compensated costs. Compensation covers medical and ancillary payments and after-tax wages net of fringe benefits (85% of gross wages). We assumed these losses were fully reimbursed for all nontrespassers, but were not reimbursed for trespassers. In computing administrative costs for deaths, we assumed only half of the lifetime after-tax wages would be replaced.
3651243 days of household production would be lost for each day of wage work lost by workers. Recall that a year of wage work averages 243 days, a year of housework 365 days. Wailer, Payne, and Skelly (1990) supports the assumption that return to labor force and household production occur on roughly the same day. Nonworkers would lose the same days of household production as workers with comparable injuries. The fringe rate for household production is 7.51%, the employer contribution to Social Security. This low fringe rate recognizes that many household laborers illegally receive no fringes or legally receive no health insurance and as part time workers receive no unemployment compensation.
RESULTS Costs by Cost Category and by Cause
Railroad injuries and illnesses caused an average of $3 billion in fatality costs and $650 million in nonfatal incident costs annually from 1987 to 1990 (in 1989 dollars). Table 1 shows the costs per case and annually for all cases. Table 2 shows similar information for nonfatal cases on1y. We also computed costs by cause group and by detailed cause. Table 3 (by cause group) implies that 85% of comprehensive injury costs in 1990 resulted from rail equipment (i.e., mainly train) crashes, rail-highway grade crossing incidents, and impacts at places other than rail-highway crossings. Maintenance of way and structures, and servicing equipment, together, accounted for 38% of injuries, but only 5% of total costs. Table 4 (by detailed cause) shows that railhighway impacts at public crossings cause at least three times as many injuries, and are three times as costly in annual terms as the next
With these assumptions, lost household production was computed in essentially the same way as lost wages. Administrative costs. For railroad workers, Railroad Retirement Board data suggest the administrative cost ratio is .104. For others, we assumed it was .118, the ratio for auto insurance (Miller et al., 1991). This value may be an underestimate since legal action is more likely when corporations with deep pockets are liable than when individuals are. Administrative costs were computed as the ratio of administrative costs to payments times
TABLE COSTS PER CASE AND DOLLARS
FOR ALL CASES OF RAILROAD
COMPREHENSIVETOTAL cases ____..__..__._.,
187.7 22.3 901.7
M M M
Per case costs include is not Included.
only cases with
cost data for all cost components.
at a 4%
. . . . . . . . . . . . . . . . . . . . . . . ,_,.__...,.., . . . . .
$33.5 M .4 M
MedIcal . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Ambulance Voc.
Wages and Fringes Household Production Administrative MONETARYSUBTOTAL Quality of Life
INJURY AND ILLNESS (1989 $1
3,652.2 M 27.482
Illness costs are lower
TABLE 2 COSTS PER NONFATAL CASE AND ANNUAL DOLLARS FOR ALL NONFATAL RAILROAD INJURY AND ILLNESS (1989 $)
cost Category MedIcal..
Voc. Rehab. Wages and Fringes Household
............................... ....................... COMPREHENSIVETOTAL
.2 M .l M
MONETARY SUBTOTAL Quality of Life
Per case costs include
is not included.
only cases with
cost data for all COST components.
at a 4%
illness costs we lower
TABLE 3 COMPREHENSIVE COSTS OF RAILROAD CASUALTIES BY CAUSE GROUP, 1990 (1990 $)
Monetan/ Injuries (Including iank 1
Injury Cause Rail equipment crossing
Group” and rail-highway
Servicing or m~intainlng eqwpment Getting on or off cars or locomotives
Flying or falling objects, burns and similar occurrences, NEC .
Operating swtches or derails Coupling or uncoupling locomotives
or on locomotive
Windows, doors, etc., equpment Coupling or uncoupling
$118,930 $109,283 $92‘61 t
air hose (or hose, and
Persons on locomotives or cars coming II? contact with fixed objects Unprovoked
turning angie cocks), safety chains
Struck by or ran into locomotives or cars at places other than rail-highway
and other crimmal
acts directed against employees and other persons not involved tn protecting 7 3
r&road property Operation of on-track
Passenger All Groups
. or mail
. . . .
’ !niurY cause codes should be distinguished h Per case refers to each injury or fatality.
from ratI equipment
Note: Each cause had a specific benefit of this detail,
Casualties from assaults which occurred I” connection with the protection of railroad property Frerght, baggage, express, Operating rail motwcars
FRA also makes
$3,961,385 use of.
for each cause W,th the
COSTS OF RAILROAD
TABLE 4 CASUALTIES BY TOP 2.5 DETAILED
I Injuries %mk 1
$ I 16,879
. .. ..
s I 12,440
along track -- equipment Casualty from reportable
moving . rail-highway
damage &5?01 . . . . . . . . Casualty from reportable rail-highway
cars not at rail-highway
moving or trestles
,....... or under cars . .
Casualty from other nonreportable rail equipment accident/incident at railhighway Railroad
crossing -- equipment moving m”tor vehicle accidents/
Casualty in reportable rail-highway impact at public crossing -- equipment standing, Casualty
monetary damage < $5701 from reportable collision of rail
other, equipment movtng Gstting on “r off cars or locomotives
Casualty in reportable ratl-highway impact at private crossing -- monetary
While working on or along track equipment moving
. .. .
other, equipment moving Crossing track not at crossing On bridges
Cost per Case
moving on-track ratl equipment, monetwv damaoe c $5701 Struck by or raiint” locomotive8
Comprehensive Cost, All Cases (Thousands)
or near track.
. . .
Sitting “r lying on track equipment moving Standing,
Monetarv Costs, Ail
Costs per Case
Casualty from reportable rail-highway impact at public crossing rnvolving moving
1990 (1990 $)
and structures falling,
-- lining switches,
on stairways, ramps, or platforms Stumbling, slipping, falling, caught, other, equipment standing
See the notes to Table
highest. Per case, the most expensive causes involve sitting, lying, standing, walking, or Winter 1994Nolume 25/Number 4
running, on or near tracks; being on bridges or trestles; and crossing tracks not at crossings. 191
Federal Highway Administration in earlier work (Miller et al., 1991; Miller, 1993). To determine costs of air crash injury, we adopted the Federal Aviation Administration’s approach (Keech, 1989). We multiplied severity-specific air crash injury counts times highway crash costs for injuries of comparable severity. All cost estimates include the costs of medical and ancillary services, emergency services, injury claims administration, and property damage, as well as lost wages, housework, and quality of life. Overall, trains are safer than highway vehicles. In terms of injury costs per 1,000 ton miles, for example, railroad freight transport is much safer than truck freight transport: the injury cost is $1.70 by railroad and $28.45 by truck (see Table 8). When viewed in terms of injury costs per vehicle mile (including passenger train miles), however, train injury costs are 25 times those for trucks because a train carries far more freight and passengers than a truck does. While fatal railroad crashes are on average double the cost of truck crashes, nonfatal railroad injuries average only 65% of the cost of highway crash injuries. For passenger transportation, Table 8 shows that commercial air travel is the safest, with
Costs by Type of Rain, and by Incident Table 5 shows railroad crash costs by type of train for 1989-90. The annual comprehensive cost of freight train crashes (here, including property damage) is $1,892 million, almost eight times higher than for passenger trains, $244 million. Per case, however, passenger train crashes are roughly twice as expensive as freight train crashes. The injury rate is twice as high. Tables 6 and 7 show crash costs by type of incident for freight and passenger trains, respectively. Crashes at rail-highway crossings account for at least 90% of passenger and freight train crash costs. Derailments, the second most common incident type, resulted in only 8% of crash costs for freight, and 6% for passenger trains. Comparing Transportation Modes Comparing injury costs among transportation modes requires examining various aspects of exposure, such as freight weight for trains and trucks, and injury costs by miles traveled for all modes. We constructed highway injury costs by vehicle type for the
TABLE 5 CRASH COSTS PER CASE BY TYPE OF TRAIN,
1989-90 (1990 $)
Freight Train Work Train
All otherd ” There
Injuries per Case
Type of Train Passenger
per year in 1989-90
$577,872 $682,413 $165,495
$37,863 $114,298 accidents
VR and ambulance),
Ifl]“ry” $80,592 $117,989
property damage and, above this level, 3062 accidents (of ten types). Cases per year does not really add up because, a defined “passenger train accident” might include a freight train, yard/switching train, etc., and vice versa. b All comprehensive
of the 3062,
short- and long-run wage and household production loss, and lost quality of life. Accident-level data contain injury and fatality w”nts, but were not linked to ~n]ury codes. To value each injury and fatality, respectively, we computed costs by person type wer all railroad injury and averaged them according to the distribution of person types in rail-equipment and rail-hlghway “accidents” and “inctdents.” ’ Property d”All other”
This was preferred damage
per case has been spread
in train accidents
by the mean in]ury cost in all railroad
the costs per m]“ry.
single cars, cut of cars, and “light locos”
SOURCE: Crash data for two calendar years, 1989-90, from the Federal Railroad Administration’s Accidentilncldent Reporting System lforms FRA F 6180.57, FRA F 6180-55, FRA F 61 SO-55a. and FRA F 6180.54) and Annual Accident/Incident Bulletins
Journal of Safety Research
TABLE 6 COMPRE?HENSIVE COSTS PER PASSENGER TRAIN INCIDENT BY TYPE OF JNCIDENT, 1989-90 (1990 $)
Cases per Year
.. .. .. . ..
Rear end Fire/violent
, .. ..
Other Obstruction Head on
. . .
Side colllsion Railroad Total
27.0000 0.4286 3.0000
$80,133 $26,591 $49,716
401.5 on Table
injury costs of $3.05 per thousand passenger miles. Trains are second-lowest at $17.60 per thousand passenger miles. Over two-thirds of this cost is attributable to injuries to pedestrians and vehicle occupants struck by trains rather than to rail passengers. Since car drivers who ignore warnings are usually at fault in rail-highway crashes, the intrinsic safety of trains relative to cars may be more striking than the table shows. Buses have injury costs of $69.35 per thousand passenger miles, still far safer than personal vehicles cars at $84.95 and general aviation at $117.25. On the other hand, when considering injury costs per vehicle mile, cars have injury costs of only $. 12 compared with commercial aircraft at $.28. Although cars have a higher
crash involvement rate, they carry fewer passengers than commercial aircraft; so less people are injured per car crash. DISCUSSlON
Data Validation and Limitations We verified the reasonableness of our medical cost estimates by comparing our average national medical cost across ICD codes (instead of railroad injury codes) to the national estimate from Rice et al. (1989; see Miller et al., 1994 for details.) After adjusting the data from Rice et al. to remove the effect of three minor assumptions that differed from
TABLE I COSTS PER F%EIGHT TRAIN ~CIDENT
and so”~ce n&es
3.5 10.5 30.5
. .. crossing .
.. .. collision .
BY TYPE OF INCIDENT,
1989-90 (1990 $)
Nonfatal Type of Incident
Cases per Year
Head on collrsion
Side collision . . . . . Rear end collision Obstruction
. .. ...,.
Other Braking collision Firelwolent rupture
0.2143 1 .oooo 0.5185
27 107 19 12
.. . . ... . ....
Broken train collision Railroad grade crossing All freight
note8 on Table
Wirlter 1994Akdume 25/Number 4
Injuries per Case
0.1362 0.0030 0.2143 0.0286 0.0790
per Case $6,759 $88.719 $432,961 $104,197 $225,765
0.0841 0.2105 0.0417
0.0000 0.0263 0.0000
$28,795 $93.089 $30,716
$370,230 $97,331 $ I .034,350 $182.007 $452,450 $177,269 $31,204 $165,133 $31,910
TABLE 8 INJURY COSTS BY MODE OF TRANSPORTATION Per 1,000 Railroad Truck Bus
.. . .,. .
.12 .28 __
117.25 of passenger
train costs resulted
crossmgs. Since train-auto crashes are usually caused by car drivers relative to cars per passenger mile may be more sulking than shown
ours, their total medical cost per case across hospital status was $725. Ours was $716 (in 1989 dollars). Although, our data sources were very different, our estimates were within 2%. The DC1 and CHAMPUS medical payments data are credible. Although our cost estimates are labeled “comprehensive,” they have limitations. They exclude the costs of travel delay, legal services, police services, and lost school days. The estimates by cost category, and by cause (Tables l-4) also exclude property damage.
car................... Commercial Air General Aviabon
who ignore warning signals, the irwtrinsic safety of train6 in this table. A similar caveat holds for freight trains.
impose large costs. Technology may offer solutions. Radar on trains is one possibility. Another is to use Intelligent Vehicle Highway Systems to reduce these crashes. ACKNOWLEDGEMENTS
This research was supported, in part, by Federal Railroad Administration contract DTFR53-91-C-00017. The authors thank Jim Blanchfield and Robert Finkelstein at FRA for their help.
The increased availability of comprehensive injury costs helps policymakers make more efficient decisions. Proposed safety interventions can be compared with themselves or with nonsafety programs, since the intangibles of death and injury are made more tangible. Injury costs are largely absorbed by the injured, through lost income and quality of life. Thus the relative safety of trains and airplanes in transport means highway modes enjoy implicit price subsidies. Better injury cost recovery through taxation or insurance would increase the price of highway travel relative to other modes. By making societal costs explicit, this would make for more informed modal choices and enhance economic e~cien~y. Despite lengthy efforts to reduce train crashes at highway grade crossings, they continue to dominate train crash costs. Since car drivers who try to “beat the train” are usually at fault in these crashes, the intrinsic safety of trains relative to cars may be more striking than we have shown. Trespasser deaths also 194
Berkowitz, M., & Burton, J. F., Jr. (1987). Permanent disabilily benefits in workers ’ compensation. Kalamazoo, MI: W.E. Upjohn Institute for Employment Research. Bureau of Labor Statistics. (1989). ~rnp~o~nzent and Earnings, 36( 1). Douglass, J. B., Kenney, J. M., & Miller, T. R. (1990). Which estimates of household production are best? Journal ofForensic Economics, 4(l), 25-46. Economic Report of the President. (1991). Washington DC: U.S. Government Printing Office. French, M. T. (1990). Estimating the full cost of workpiace injuries. American Journal ofPublic Health, 80, 1118-1119. French, M. T., & Kendall, D. L. (1992). The value of job safety for railroad workers. Journal of Risk and Uncertainty, 5, 175-185. Hirsch, A. E., Eppinger, R. H., Shame, T., Van Nguyen, T., Levine, R. S., Mackenzie, J., Marks, M., & Ommaya, A. K. (1983). ~Fnp~irrnent stating from the abbreviated injury scale. Washington, DC: Federal Highway Administration. Keech W., Hoffer, S., Timmerman, C., & Bernadino, F. (1989). Economic values for valuation of federal aviation administration investment and regulatory programs (Report FAA-APO-220). Melinek, S. J. (1974). A method of evaluating human iife for economic purposes. Accident AnaEysis and Prevention, 6, 103-114.
of Safety Research
Miller, T. R. (1993). Costs and consequences of U.S. roadwav crashes. Accident Analysis and Prevention, 25(5), 593-607. Miller. T. R.. Viner. .I. G.. and Associates. (1991). The costs of highway crashes. Washington, DC: Federal Highway Administration. Miller, T. R. (1990). The plausible range for the value of life: Red herrings among the mackerel. Journal of Forensic Economics, 3(3), 17-40. Miller, T. R., Pindus, N. M., Douglass, J. B., & Rossman, S. B. (accepted for publication, 1994). Nonfaral injury incidence, costs, and consequences: A data book. Washington, DC: The Urban Institute Press. Miller, T. R., Calhoun, C., & Arthur, W. B. (1989) Utilityadjusted impairment years: A low-cost approach to morbidity valuation. In Estimafing and vnluing morbidity in a policy context: Proceedings of a June
1989 AERE Workshop (EPA-230-08-89-065). Washington, DC: U.S. Environmental Protection Agency. Rice, D. P., MacKenzie, E. J., and Associates. (1989). Cost of injury in the United States: A report to Congress. San Francisco, CA: Institute for Health and Aging, University of California, and Injury Prevention Center, Johns Hopkins University. Sackett, D. L., & Torrance, G. W. (1978). The utility of different health states as perceived by the general public. Journal of Chronic Disease, 31,697-704. U.S. Office of Management and Budget. (1989). Regulatory program of the United States. Washington, DC: U.S. Government Printing Office. Waller, J. A., Payne, S. R., & Skelly, J. M. (1990). Disability, direct cost, and payment issues in injuries involving woodworking and wood-related construction. Accident Analysis and Prevention, 22(4), 35 I-360.