Professional Documents
Culture Documents
SAE TECHNICAL
PAPER SERIES 2000-01-1354
400 Commonwealth Drive, Warrendale, PA 15096-0001 U.S.A. Tel: (724) 776-4841 Fax: (724) 776-5760
Downloaded from SAE International by University of British Columbia, Monday, July 30, 2018
The appearance of this ISSN code at the bottom of this page indicates SAE’s consent that copies of the
paper may be made for personal or internal use of specific clients. This consent is given on the condition,
however, that the copier pay a $7.00 per article copy fee through the Copyright Clearance Center, Inc.
Operations Center, 222 Rosewood Drive, Danvers, MA 01923 for copying beyond that permitted by Sec-
tions 107 or 108 of the U.S. Copyright Law. This consent does not extend to other kinds of copying such as
copying for general distribution, for advertising or promotional purposes, for creating new collective works,
or for resale.
SAE routinely stocks printed papers for a period of three years following date of publication. Direct your
orders to SAE Customer Sales and Satisfaction Department.
Quantity reprint rates can be obtained from the Customer Sales and Satisfaction Department.
To request permission to reprint a technical paper or permission to use copyrighted SAE publications in
other works, contact the SAE Publications Group.
No part of this publication may be reproduced in any form, in an electronic retrieval system or otherwise, without the prior written
permission of the publisher.
ISSN 0148-7191
Copyright 2000 Society of Automotive Engineers, Inc.
Positions and opinions advanced in this paper are those of the author(s) and not necessarily those of SAE. The author is solely
responsible for the content of the paper. A process is available by which discussions will be printed with the paper if it is published in
SAE Transactions. For permission to publish this paper in full or in part, contact the SAE Publications Group.
Persons wishing to submit papers to be considered for presentation or publication through SAE should send the manuscript or a 300
word abstract of a proposed manuscript to: Secretary, Engineering Meetings Board, SAE.
Printed in USA
Downloaded from SAE International by University of British Columbia, Monday, July 30, 2018
2000-01-1354
1
Downloaded from SAE International by University of British Columbia, Monday, July 30, 2018
H ( ∆v k ) : Average harm per collision resulting in REAR-END COLLISION – In a rear-end collision, the
∆vk for a specific manner of collision (e.g., post-crash speed is expressed mathematically as [7][8]:
rear-end impact).
( m F × v cF + m L × v cL )
N (S , ∆v k ) vf = (11)
Pw (Si , ∆v k ) = w i (9) (m F + m L )
N w (S i )
mF: Mass of following vehicle.
Nw(Si,∆vk): Number of relevant collisions that were
mL: Mass of lead vehicle.
preceded by a safety-critical driving con-
flict Si and resulted in ∆vk with the assis- vcF: Velocity of following vehicle at time of impact.
tance of an IVSS. vcL: Velocity of lead vehicle at time of impact.
Nw(Si): Number of relevant collisions that were vf: Final velocity of the coupled vehicles after
preceded by a safety-critical driving con- impact.
flict Si with the assistance of an IVSS.
The changes in speed as a result of the impact for the fol-
N wo (S i , ∆v k ) lowing vehicle, ∆vF, and for the lead vehicle, ∆vL, are then
Pwo (S i , ∆v k ) = (10) computed by:
N wo (S i )
Nwo(Si,∆vk): Number of relevant collisions that were ∆vF = vcF – vf ∆vL = vcL – vf (12)
preceded by a safety-critical driving con-
flict Si and resulted in ∆vk without the LANE CHANGE COLLISION – Figure 1 illustrates one
assistance of an IVSS. vehicle (Number 1) steering toward another vehicle
(Number 2) that is traveling straight in an adjacent lane,
Nwo(Si): Number of relevant collisions that were
prior to a lane change collision.
preceded by a safety-critical driving con-
flict Si without the assistance of an
IVSS. y
3
Downloaded from SAE International by University of British Columbia, Monday, July 30, 2018
vfy: Lateral component of final velocity of coupled VEHICLE TYPE AND WEIGHT DISTRIBUTION
vehicles after impact:
Distributions of vehicle types and weights on US road-
m × v1c × sin( θ)
vf y = 1 (16) ways are needed to compute vf as indicated in Equations
(m1 + m 2 ) (11), (14), and (16). Figure 2 shows the distribution of
vehicle types based on 1996 US registration statistics
In a lane change collision, total ∆v for vehicle 1, ∆v1, and
that counted approximately 203 million registered vehi-
total ∆v for vehicle 2, ∆v2, are then calculated as follows:
cles [9]. The “truck” type encompasses light trucks (≤
4,500 Kg), medium trucks, and heavy trucks (> 15,000
∆v1 = ∆v12x + ∆v12y ∆v 2 = ∆v 2 2x + ∆v 2 2y (17) Kg). Light trucks accounted for about 93.3 percent of all
trucks in 1996. Figure 3 illustrates the distribution of light
SINGLE VEHICLE ROADWAY DEPARTURE (SVRD) vehicles, autos plus light trucks, which comprised about
COLLISION – The total ∆v in SVRD collision type may be 95.3 percent of all registered vehicles in 1996. Table 2
expressed as: indicates that the average weight of automobiles varied
between 1,139 Kg for subcompact autos and 1,657 Kg
∆v = α × vc (18)
for large autos over 1992-1997 period. The distribution of
α: Impact severity attenuation factor. the truck population by gross vehicle weight, excluding
vc: Speed of vehicle at time of impact. light trucks, is shown in Figure 4 based on 1992 statistics.
11.4
Table 1. Distribution of “Objects Collided With” in SVRD
8.9
Crashes
Object %*
Fixed 30.1 1.2 0.8
Breakaway 24.3 Truck <= 2700 Kg Compact Midsize Subcomp Truck<= 4500 Kg Large 2-Seaters Minicompact
4
Downloaded from SAE International by University of British Columbia, Monday, July 30, 2018
7.6 The 1994 and 1995 CDS crash databases were queried
4.5 4.3
2.8
to estimate the values of H o ( ∆v k ) for rear-end, lane
"4.5 – 6.4" "6.4 - 7.3" "7.3 – 8.8" "8.8 – 11.8" "11.8 - 15" ">15" change, and SVRD collisions. A total of 1,235 crash
Thousand (Kg)
cases or about 28 percent of all examined cases con-
tained valid ∆v information. Each crash case was
Figure 4. Distribution of Medium/Heavy Trucks by
weighted using its Ratio Inflation Factor in the CDS so as
Weight Based on 1992 Statistics
to approximate the national profile. Moreover, injuries are
scored in the CDS database using the police-reported
AVERAGE HARM PER COLLISION
injury score KABCO which is converted to the MAIS
scale by a translator derived from the 1982-1986 National
The average harm per collision resulting in ∆vk, H ( ∆v k ) Accident Sampling System (NASS) data.
in Equation (8), is obtained for rear-end, lane change,
and single vehicle roadway departure (SVRD) collisions Table 4. Average Values of Vehicles Involved and
in 10 Km/h speed bins as follows: Vehicle Occupancy in Collisions^
N0(∆vk):Number of occupants of a specific crash type ^: Based on annual statistics from 1989-93 GES Crashes [11].
resulting in ∆vk.
w(i): Weighting coefficient corresponding to maxi- AVERAGE HARM PER OCCUPANT IN REAR-END
mum injury severity i based on the Maximum COLLISIONS – A total of 455 rear-end crash cases or
Abbreviated Injury Scale (MAIS). about 38 percent of all rear-end crash cases in the 1994
I(i): Total number of persons involved with a max- and 1995 CDS databases had valid total ∆v information
imum injury severity i based on MAIS result- for both the lead and following vehicles. These cases
ing from ∆vk collisions.
were examined to estimate H o ( ∆v k ) based on the high-
est total ∆v experienced in each collision. Figure 5
5
Downloaded from SAE International by University of British Columbia, Monday, July 30, 2018
illustrates the relationship between ∆v and average harm Fatal Equivalent Value
0.035 0.034 0.034 0.034
per occupant in rear-end collisions, expressed in terms of 0.032
fatal equivalent values. The average harm per occupant 0.03
should increase with increasing values of ∆v. However, 0.024 0.025
0.025
there appears to be an anomaly in the distribution due to
the low number of cases available in some ∆v bins, espe- 0.02 0.017
cially in the “15 – 25” and “> 65” Km/h ∆v bins. It is rec- 0.015 0.012
ommended that a fatal equivalent value of 0.028 be
0.01 0.008
assigned to H o ( ∆v k ) at ∆v above 55 Km/h (using the 0.005
0.005
weighted average of fatal equivalent values at the last two
0
∆v speed bins). Moreover, a value of 0.006 is suggested "5 - 15" "15 - 25" "25 - 35" "35 - 45" "> 45"
for the “15 – 25” Km/h ∆v bin based on the increasing ∆v (Km/h)
rate of H o ( ∆v k ) between “25 –35” and “45 – 55” Km/h CDS Modified
∆v bins. Figure 6. Distribution of Average Harm Per Occupant in
Lane Change Collisions by Highest ∆v
AVERAGE HARM PER OCCUPANT IN LANE CHANGE
COLLISIONS – The 1994 and 1995 CDS crash data- AVERAGE HARM PER OCCUPANT IN SVRD
bases contained 67 cases or about 22 percent of all lane COLLISIONS – A total of 713 cases or about 24 percent
change crash cases with valid total ∆v information for of all SVRD crash cases with valid total ∆v information
both vehicles involved in the collision. The values of were extracted from the 1994 and 1995 CDS databases.
H o ( ∆v k ) are estimated using the highest total ∆v expe- Figure 7 illustrates the distribution of the average harm
rienced by either vehicle in each collision. Figure 6 illus- per occupant in SVRD collisions by 10 Km/h ∆v bins. The
trates the relationship between ∆v and average harm per average harm per occupant is expected to increase with
occupant in lane change collisions as estimated from the ∆v as evident in Figure 7. This harm increased dramati-
CDS database and displays a modified distribution to cally between “45 – 55” and “55 – 65” Km/h ∆v bins, and
adjust the observed anomaly. As seen in Figure 6, an then was reduced slightly at ∆v over 65 Km/h. This may
anomaly exists in the distribution of observed injury be attributed to the small number of valid cases available
severity data due mainly to the very low number of lane for the last two speed bins. It is recommended that a fatal
change crash cases available in the two-year period equivalent value of 0.376 be assigned to H o ( ∆v k ) at ∆v
under consideration.
above 55 Km/h (using the weighted average of H o ( ∆v k )
Fatal Equivalent Value at the last two ∆v bins).
0.041
0.04 Fatal Equivalent Value
0.4 0.389
0.359
0.03
0.024
0.3
0.02 0.017
0.015 0.014
6
Downloaded from SAE International by University of British Columbia, Monday, July 30, 2018
SUMMARY AND CONCLUDING REMARKS 2. National Center for Statistics and Analysis, “National
Automotive Sampling System, 1995 Crashworthiness
This paper delineated a novel methodology and funda- Data System, Data Collection, Coding and Editing
mental supporting data to assess the impact of intelligent Manual”. United States Department of Transporta-
tion, National Highway Traffic Safety Administration,
vehicle safety systems on the number of collisions and
Washington, D.C., 20590.
the number and severity of crash-related injuries using
computer simulations. Simple mathematical models with 3. Najm, W.G., Wiacek, C.J., and Burgett, A.L., “Identifi-
representative parameters were described which derive cation of Precrash Scenarios for Estimating the
Safety Benefits of Rear-End Collision Avoidance Sys-
the parameter ∆v from speeds at impact and weights of
tems”. Fifth World Congress on Intelligent Transport
vehicles involved in rear-end, lane change, and SVRD Systems, Seoul, Korea, October 1998.
collisions. The SVRD collision model requires data on the
impact severity attenuation factor, α, which are not readily 4. NHTSA Benefits Working Group, “Preliminary
Assessment of Crash Avoidance Systems Benefits”.
available. Data are needed to better describe α for differ-
Version II, United States Department of Transporta-
ent object categories. The 1994 and 1995 crash data- tion, National Highway Traffic Safety Administration,
bases were queried to identify the relative frequency of December 1996.
striking five object categories in SVRD collisions. In sup-
5. Najm, W.G. and Burgett, A.L., “Benefits Estimation
port of ∆v estimation, statistics were presented on vehicle for Selected Collision Avoidance Systems”. Fourth
types and weights based on 1996 vehicle registration in World Congress on Intelligent Transport Systems,
the U.S. Berlin, Germany, October 1997.
Statistical data on the number of vehicles involved per 6. Campbell, K.L., Joksch, H.C., and Green, P.E., “A
crash and the number of occupants per vehicle were pro- Bridging Analysis for Estimating the Benefits of
vided in order to estimate the average harm per collision. Active Safety Technologies”. NHTSA Contract No.
The 1994 and 1995 CDS crash databases were exam- DTNH22-93-D-07000, UMTRI-96-18, April 1996.
ined to obtain relationships between ∆v and average 7. Evans, L., “Driver Injury and Fatality Risk in Two-Car
harm per occupant in each of the three collision types. Crashes Versus Mass Ratio Inferred Using Newto-
Unfortunately, a small number of crash cases in the CDS nian Mechanics”. GMR-7928, April 1993.
database contained valid information on ∆v. Conse- 8. Khadilkar, A.V., Redmond, D., and Ausherman, V.K.,
quently, the results obtained had anomalies in the distri- “Collision Avoidance System Cost-Benefit Analysis”.
bution of the average harm per occupant with respect to Volume I, United States Department of Transporta-
∆v. To remedy this problem, at least 5 years of CDS data tion, National Highway Traffic Safety Administration,
might be needed to generate the appropriate distribu- DOT HS 806 242, September 1981.
tions. 9. Davis, S.C., “Transportation Energy Data Book: Edi-
tion 18”. ORNL-6941, U.S. Department of Energy,
ACKNOWLEDGMENT DE-AC05-96OR22464, September1998.
10. Blincoe, L.J., “The Economic Cost of Motor Vehicle
Mr. John Smith of the Volpe Center’s Accident Prevention Crashes, 1994”. United States Department of Trans-
Division is recognized for his contribution to this research portation, National Highway Safety Administration,
DOT HS 808 425, July 1996.
work by providing data from the 1994 and 1995 CDS
crash databases. 11. Wang, J.-S., Knipling, R.R., and Blincoe, L.J., “Motor
Vehicle Crash Involvements: A Multi-Dimentional
Problem Size Assessment”. ITS America Sixth
REFERENCES
Annual Meeting, Houston, TX, April 1996.
1. Najm, W.G. and daSilva, M.P., “Benefits Estimation 12. Knipling, R.R., Mironer, R., Hendricks, D.L., Tijerina,
Methodology for Intelligent Vehicle Safety Systems L., Everson, J., Allen, J.C., and Wilson, C., “Assess-
Based on Encounters with Safety-Critical Driving ment of IVHS Countermeasures for Collision Avoid-
Conflicts”. Submitted for Publication at Intelligent ance: Rear-End Crashes”. Appendix B, United
Transportation Society of America’s Tenth Annual States Department of Transportation, National High-
Meeting & Exposition, Boston, MA, May 2000. way Traffic Safety Administration, DOT HS 807 995,
May 1993.