Nr 62 ' 1981
62
SARBMCEE
Statens väg- och trafikinstitut (VTI) - 581 01 Linköping
National Road &Traffic Research Institute - S-581 01 Linköping - Sweden
Daylight Running Lights in Sweden
Pre Studies and Experiences
by Kåre Rumar
Paper presented at the 1981 SAE Congress, Detroit, Michigan, USA,
February 23-27, 1981.
'
Published as SAE Technical Paper Series No. 810191, Society of Automotive
...
-.-.-.;.;.;.;...
810191
Daylight Running Lights
in Sweden Pre-Studies
and Experiences
Kåre Rumar
National Swedish Road and Traffic Research Institute (VTI)
International Congress and Exposition
Cobo Hall, Detroit, Michigan
February 23-27, 1981
'.'.'.'. 51.51.1- ;.;.;.;._._ _______ . ...75th
anniversaryThe appearance of the code at the bottom of the first page of this paper indicates SAE s consent that copies of the paper may be made for persona! or internal use, or for the personal or internal use of specific clients. This consent is given on the con-dition, however, that the copier pay the stated per article copy fee through the Copyright Clearance Center, Inc., Operations Center, P.O. Box 765, Schenectady, NY. 12301, for copying beyond that permitted by Sections 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.
Papers published prior to 1978 may also be copied at a per paper fee of $2.50 under the above stated conditions.
SAE routinely stocks printed papers for a period of three years following date of publication. Direct your orders to SAE Order Department.
To obtain quantity reprint rates, permission to reprint a technical paper or per-mission to use copyrighted SAE publications in other works, contact the SAE Publications Division.
ISSN 0148-7191
SINCE MOST OF THE INFORMATION ACQUISITION in
road traffic is made by the visual channel the correlation between accidents and visual perform ance of involved drivers should be high.
Further-more the most common explanation that is heard
at traffic courts as explanation to accidents is "I did not see him in time". It is therefore amazing that in spite of numerous studies only very low correlations (r<.l) are reported
(1, 2)x. One explanation is that we are not
studying the relevant visual functions. What then are the relevant visual functions? During the
history of man we have been both hunted and
hunters. In those two roles the detection of moving creatures (attacking or escaping) in the peripheral visual field was a matter of life and death. And according to Darwin what was import ant to survival soon became a characteristic feature of the species. The attacking and escaping animals in the history of man were really moving jumping, running, flying etc. However, in present road traffic the "attacking vehicles are not really moving. They are silently and without any inherent motion sneaking up on us at a very high speed. Therefore this inherited talent of ours does not function in the fight
ABSTRACT
810191
Daylight Running Lights
in Sweden Pre-Studies
and Expe ences
Kåre Rumar
National Swedish Road and Traffic Research Institute (VTI)
for survival in traffic. We have to replace the historic motion by something that gives the same
type of peripheral conspicuity to vehicles. This is in fact the general background to the intro-duction of daylight running lights in Sweden.
CONSPICUITY
From psychophysical point of view the detection of a visual target is determined by its
contrast against background (brightness and colour)
angular size motion
The size of a vehicle cannot be increased just to increase its conspicuity. Some
sugges-tions have been put forward how real motion
could be introduced to vehicles - e.g. on the front of trains to scare reindeers in the
northern areas where this is a real problem. XNumbers in parantheses designate References at
end of paper.
The usage of daytime running lights in Sweden started at the end of the 60:ies. The behavioural studies of various measures to improve vehicle daylight conspicuity started at the same time. The conspicuity of oncoming vehicles was analyzed for drivers in real traffic. Brightness and colour contrast were found to be the most common causes to detection. The effect of different vehicle colours and lighted low beams on detection time was studied.
0148-7191/81/0223-0191$02.50
The low beam condition was in all situations as good as the best colour. Many field experiments of peripheral detection as a function of running light intensity were carried out. Various types of running light systems were developed. The costs of a running light were calculated. A running light specification (intensity: 300 800
cd) and a law concerning its use (1 Oct. 1977)
were written. The accident analyses indicate a very favourable effect of daylight running lights on collision accidents in daylight.
But for motor vehicles on the roads this is no
realistic method, Consequently, what remains
to manipulate is contrast. A considerable num
ber of studies have been carried out to study
the contrast sensitivity of the human visual system. Blackwell (3) is the largest and most
well known. What has been done in the Scandina
vian countries is mainly to study directly or indirectly the safety effects of changes in
contrast between vehicle and background. The lower the illumination level the higher the
contrast has to be to reach the same conspicuity.
There are special Nordic reasons for trying to
increase the contrast between vehicle and back
ground. We have a generally lower level of am
bient illumination due to comparatively low altitude of the sun. In December the sky illu mination in Washington is five times that of
Stockholm. In June the dawn and dusk periods in Stockholm are about three times as long as in
Washington. The proportion of overcast daylight hours per year is in Washington 432 and in Stock holm 562. The effect of overcast is larger when the sun is lower. Low standing sun is also a
considerable glare source that may decrease contrasts in the Visual field (4).
NORDIC STUDIES
As mentioned before the studies of safety
can be direct (effects on accidents) or indirect
(behavioural). The direct methods seem very
accurate with several decimals in the figures.
But the results have to be treated with great caution. Accidents as a criterium is not too reliable. There are a great many factors behind an accident, and it is really difficult to isolate the effect of a single cause.
The indirect (behavioural) studies can be
split up into
calculations based on laboratory data - laboratory experiments
- full scale experiments - real traffic observations
All four types have been used in the Nordic studies. The drawback with the indirect studies are that the obtained effect is very difficult to translate into real safety figures.
First an account is given of the use of lights on during daylight hours in Sweden before the law. Thereafter the behavioural studies are described. Then the few studied technical and economical aspects are reviewed. Finally the results of the accident analyses, the usage after the law and some practical experiences are
reported.
PRELAW USE OF DAYTIME LIGHTS ON
Sweden changed over from left to right hand
traffic in 1967. In that specific situation it was considered important to have high peripheral conspicuity on vehicles since road users would probably often look in the wrong direction.
810191
Therefore the use of low beam during daylight was promoted by road safety organisations.
How-ever, the usage was never very high - only about ZZ in clear weather (5). In the summer of 1974
the usage in clear weather was up to 10%, in the autumn of 1976 the corresponding figure was 25%. At that time the official promoting cam paign was quite intensive and both Volvo and Saab were equipping their new passenger cars with standard running lights (see below). The
figures were of course very dependent on the level of ambient illumination. In figure 1 the proportion of vehicle lighting is given as a function of weather condition and sky illumina tion in Sweden the winter 1975/76 (6). Just
before the law (1977) the usage was about 50%.
The Swedish Railways introduced daytime lights on all their locomotives already in 1966/67. In 1967 the Swedish Police made low beam during daylight hours compulsory on all their cars except for cars on special missions. The Swedish Defence Forces introduced compul sory low beam on all their vehicles in 1969 (with the exception for war and manoeuvresf). Several separate trucking and bus companies introduced the same rules at the end of the 60:ies. All experience reported was in favour,
USAGE (%)
100
KRAlN,SNOW
HAZE,FOG
___OVERCAST
50
10
I I10
20
30
SKY ILLUMINATION (103LUX)
N
Fig. 1 Usage (%) of daytime running lights in Sweden 1975/76 as a function of sky
810191
and those who introduced the rule kept it until the law was introduced in 1977. However, no real
follow up studies were made not of effects on accident nor of effects on vehicle costs.
Official and other road safety organisations, the large motorist organisation and a majority of the public were during this prelaw period very positive to the concept of lights on during daytime in order to increase vehicle conspicuity. One prominent ophthalmologist (7) and several other individuals thought the idea
completely crazy dangerous due to glare and very costly due to increased petrol and bulb
consumption.
In Finland the situation was comparable up to 1968. After that the Finnish authorities moved faster concerning daytime lights on and already in 1969/70 the usage was as high as about 50%. Norway on the other hand has moved slower and in spite of some official campaigns the usage 1975/76 was only about ZZ. In Denmark no official campaigns were carried out and the usage in 1975/76 was very low.
BEHAVIOURAL STUDIES
At the end of the 60:ies Dahlstedt & Rumar
(8) carried out a study of causes for the de
tection of oncoming vehicles in rural traffic. Colour slides and verbal descriptions were taken at the moment of detection in real traffic.
Their results show that brightness contrast
(colour, silouette, flashes, headlights) is the dominating single factor (> 80%) both during summer as well as winter. Colour contrast
occurred in some cases at high levels of general illumination ( 15%). Motion was the smallest
single cause (5 10%). When headlights (low beam) were on this was invariably the cause for detection.
On the basis of these results Dahlstedt &
Rumar made some laboratory studies of the
con-spicuity of 16 various car colours with and without headlights against some common traffic backgrounds. The subjects were to detect "the vehicle" on a colour picture projected for a very short time in front of them. Detection time was the criterium. The results show that optimal colour for conspicuity changed with
background. But as soon as the background was
not the sky or a snow field, headlights on gave
the same effect as the best colour. After these results the study of car colour was abandoned,
and the following studies of vehicle conspicuity were made with lights.
Rumar and Hörberg have in several studies
(9, 10, 11, 12) systematically investigated the central and peripheral conspicuity of oncoming vehicles as a function of vehicle lighting intensity, surface and colour and as a function of level of ambient illumination and peripheral angle. Other smaller studies with related pur poses are also reported (13, 14, 15).
The results show that central vision vehicle conspicuity is improved even by very weak lights (N 50 cd). But as Rumar argues (12) it is the peripheral conspicuity that is im-portant. It is in situations when a vehicle appears where it is not expected (fixated) that the issue may be critical that the situation may develo to an accident.
In 60 peripheral vision a considerably higher intensity (> 400 cd) is needed to improve vehicle conspicuity between 3 000 and 6 000 lux sky illumination. At 300, an intensity of 400 cd almost doubles the detection distance of a car compared to the same car without lights (see figure 2). At 200 the detection distances are improved by running lights of 100 - 300 cd at about l 000 lux sky illumination or lower (see figure 3).
No real difference between lighting colour (white and yellow) and lighting surface (70 and 200 cm ) was obtained. A correctly aimed standard European low beam gives 350 450 cd into the eyes of an oncoming driver on a flat straight
road. On uneven roads, curves, hillcrests, by
misaimed headlights etc., the intensity levels might be several times higher. No real differ ences in estimated distance to vehicles with and without low beam as observed in rear view mirrors were reported (14).
Hisdal (16) has tried to calculate on the
basis of the contrast sensitivity of the eye
DETECTION
DISTANCE (m)
800-d 600-#RUNNING LIGHT INTENSITY
LOG cd
j_
CT
U
L ||so $050
66000
A 70cmz SURFACE
x 200 cm2 SURFACE
a NO LIGHT ON THE CAR
Fig. 2 Detection distances (m) of ongoming vehicles for two peripheral angles (30 , 60 ) as a function of running light intensity (cd)
600 A RUNNING UGHT INTENSITY F «OCD 1x DwainCD
5°°~ axx QX
oo 200c0
__
cu
Air-A 300CD
E \ \ . LLJ \'\ X _ z 400- xxx 1 F \\\ 1 w 1 A éiymr % gF:
&
\ 8 *$( A MI-50k .r_L 1 l l 1 A 0 125 350 625 025 1250 1750SKY ILLUMINATION (LUX)
Fig. 3 Peripheral (200) detection distances (m) of oncoming vehicles as a function of sky
illumination (lux) and running light intensity
(cd)
(3) the probability to miss during a one second observation at a distance of 100 m, in dawn/dusk illumination, an oncoming vehicle with and without low beams. According to Hisdal the risk
for a miss would be about lZ for an oncoming vehicle without light and almost none for a vehicle with low beams.
One of the arguments against daytime lights on has been the possible masking of braking lights by the rear position lights. In a Swedish study (17) this was tested at peripheral angles
2 , 5 and 100 with reaction time as a criterium and no masking effect was found.
In the discussion of suitable intensity levels of a daylight running light the possible use of the running light during darkness within lighted areas (city beam) has been a complica-tion. Hörberg & Rumar (9) argue for a daytime intensity in the interval 1 000 500 cd a d a nighttime intensity in the interval 100
-50 cd. They suggest an area of at least 70 cm
and a yellow or white colour. In their later
study (11) they seem to favour a lower inten sity. Hisdal (18) considering a double use
(day and night) argues for an intensity of about 1/5 of the low beam. Hörberg & Rumar (9) suggest as a compromise day ånd night intensity ~ 200 cd
and an area of 50 cm . RUNNING LIGHT ALTERNATIVES
The daytime running light alternatives that have been discussed in Sweden are standard low beam (incandescent or halogen), special running lights, fortified position lights,
re-duced low beam, rere-duced high beam, standard or
reduced front curve and fog head lights, cor nering lights. The standard low beam has always been and will probably in the future be accepted.
810191
The other solutions are still under discussion
but accepted.
Special running lights to place on or under
the front bumper are presently mounted on about
152 of the cars. After a long discussion a Swe dish standard mainly dealing with light distri-bution was accepted in 1978 (19). The standard
(see figure 4) corresponds to ECE Regulations
on e.g. front position lights but the illumi-nating surface should be ) 40 cm and the central intensity should be > 300 cd and S 800 cd. The colour could be white or yellow but is in practice yellow. These special running lights cost about $ 20 and normally use 21 w long life
lamps.
Fortified parking lights was the solution chosen by Volvo when they in 1975 introduced the running light as standard equipment on their
passenger cars. The bulb has two filaments 5 w
(position light) and 21 w running light:(both filaments). This running light gives 300 400 cd straight ahead and has very good conspicuity at
large angles.
Reduced low beam was the initial solution
of Saab when they introduced their standard
running lights in 1975. A resistor of about .4Q is inserted in the circuit and this reduces the luminous output with about 50%. The life time of the bulb will be about ten times as long (4). Initially there was a fear that since lowered voltage destroys the halogen process in the bulb
this would create a problem. However, both laboratory and field experiences show that even
if some blackening of the bulb takes place this is reversible and disappears when full voltage is applied again.
Reduced high beam has been sold to a small
extent as auxiliary equipment. As mentioned
above the halogen bulb does not appear to be
any problem. But the conspicuity at large angles and the reddish colour due to large voltage reduction are problems. This solution will probably be prohibited in Sweden.
+ 1001
60 V 60
+ 50- 30
60
Zlo
60
30
0°~ H
105 270 300 270 105
H
5°~~ -3p
ago
jo
6:0
3io
-1004L %
% 6|0
60
:
I
|
|
I
|
I
|
I
0° +5° +10°
+20O
Fig. 4 - Minimum light distribution of special running lights according to Swedish standard
810191
Standard and reduced curve lights have not
really been used to any extent. Cornering lights combined with running lights and parking lights were introduced by Saab as a standard in 1977.
RUNNING LIGHT COSTS
The costs for various alternatives of
day-light running day-lights have been calculated in various ways. The method used in Sweden (20) was to split it up into installation costs (old
and new cars) and running costs (bulbs and
petrol).
The calculations of installation costs gave the following results ($).
new old
cars cars.?
manual
Standard low beam automatic 3 20 Reduced low beam manual 7 30 (12V+1OV) automatic 8 , 35
Special running lights automatic 25 40 The theoretical calculations of yearly running cost are based on an increased burning time of 150 hours, the simultaneous use also of
the other lamps on the passenger car, a petrol
price of about 50 cents per liter and a petrol consumption of .09 litres for production of 100 w electrical effect per hour.
The results were the following ($).
Bulbs Petrol Z
Incandescent! 4 5 9
Standard low beam Halogen 8 6 14
Reduced low beam Incandescent 1 4 5 (12V+10V) Halogen 2 5 7
Spec. running lights Incandescent 2 3 5 Special empirical tests of petrol consump
tion with and without running lights (21) re
sulted in a difference of about 12.
Using an expected life time of the car of
14 years, an average driving distance per year,
and an interest of 82 makes it possible to
capitalize the costs as follows (in figures for
comparison).
Standard low beam Incandescent 100 (norm)
Halogen 150
Reduced low beam 80
Spec. running lights 116
As can be seen the reduced low beam is the
economically most favourable solution.
The conclusion is that the introduction of a compulsory daylight running light would result in an increase of automobile costs for the in dividual car owner of about 12. It should be remarked that critics of daytime running lights
have presented calculations that are considerably larger.
ACCIDENT STUDIES
Some studies of the effect of daytime running lights on accidents have been carried out in the Nordic countries. The first one carried out by the Swedish Road & Traffic
Re-search Institute (VTI) in cooperation with
Finnish authorities (22) concerns the Finnish situation 1968 - 1974. From 1968 to 1970 there was a general campaign for daytime low beam. This resulted in a usage of about 502. 1970 an official recommendation was issued to use low beams outside built up areas from October 1 to March 31. The usage increased to about 852. Finally, from 1972 1974 (and still) the use of low beam during the same time and the same area
was compulsory. The usage now went up to over
952.
The hypothesis tested was that the greater use of low beams should reduce multiple accidents in daylight but leave single accidents in day light and both multiple and single accidents in
darkness unaffected.
A linear regression analysis of the accident development from 1968 1974 gave the following results:
Multiple accidents in daylight (MD) decreased by 322
- Multiple accidents at night (MN) decreased by 42
- Single accidents in daylight (SD) decreased by 42
Single accidents at night (SN) increased by 62
In an effort to balance the possible effect
of other measures than daytime low beam the ratio A was formed
MD
W
SD
&
A=
Standardized this way
1nu1tip1e accidents in daylight decreased by
152 from the campaign to the recommendation
period and by a further 62 to the compulsory period.
It should be noted that this large effect was obtained by an increase of daytime low beams from about 502 (not from 02) to 952.
The Swedish law concerning daytime running lights was introduced October 1, 1977 for cars and motorcycles and 1980 for tractors, working
machines, and mopeds. The law was general - that
is to say concerned all vehicles at all times everywhere in Sweden.
In a first study of the effect of the day time running light law on accidents in Sweden the preperiod 1975 - 1977 was compared to the
afterperiod 1977 - 1978 (23). The same
statis-tical technique was used as in the Finnish
study (22). However, this time it was possible to separate between rural and urban areas,
between winter and summer and between some
weather conditions.
The change in usage between the prelaw and postlaw periods varies a great deal with time of
year and weather condition. The usage increased
in clear weather during winter from 55% to 95% in clear weather during summer from 25% to 90% in overcast during winter from 80% to 95% in overcast during summer from 35% to 95% in snow or rain during winter from 90% to 100% in rain during summer from 70% to 95%
The results based on the same type of ratio (A)as in the Finnish study were the following (none of them statistically significant).
Multiple accidents in daylight decreased by 20% during winter in urban areas
by 17% during winter in rural areas by 12% during summer in urban areas
increased by 3% during summer in rural areas the weather conditions gave no clear cut
results
The winter effect seems quite stable while the summer effect shows considerable variation. Again it should be noted that these results were obtained by comparison between a high usage (roughly 50%) and a very high usage (about 95%). Both the usage figures and the obtained effects on accidents coincide well with the results from the Finnish study.
Presently, also the effect during the se
cond year after the law (1978 1979) is stu died. Some modifications of statistical analyses have been introduced. The preliminary results
(24) indicate lower but more stable effects. The Swedish Road Safety Office (TSV) has
also studied the effect of the running light law on Swedish accident statistics (25). The
method corresponds closely to the VTI method,
but as a control TSV has been using rear end accidents instead of single accidents. They have also split up the multiple accidents into various types (overtaking, turning, oncoming etc), fi
nally they have been using a longer prelaw pe
riod (1970 76). This technique has given larger and significant results in favour of day time running lights.
The conclusion is that daytime running lights reduce daytime collision between motor vehicles and other types of road users. The size of the reduction seems to be 5 15%.
GENERAL EXPERIENCES
The public attitude to a running light le gislation the year before the legislation was very good (26). Among those already using day time lights on, about 90% were in favour, among those not using lights, 30 40% were in favour. The number of young drivers and drivers driving long distances were overrepresented in the lights on group. After the law the general atti tude is still very good. An indication of that _is the still very high usage close to 95%
810191
except for large cities where the usage is
about 85% (27).
The critisism has come from bus and truck companies who claim that the increased costs
due to increased bulb exchanges and petrol con
sumption are substantial since they have a
large number of vehicles. One of the reasons
for their critisism is that at the time of in-troduction only the front lights were compul sory. Due to misuse in dawn and dusk periods
this was later changed and now also the rear
lights are compulsory. Especially for trucks with trailers and side marker lights this makes
a large difference.
In Sweden headlight cleaners are compulsory on cars built 1976 or later. This means that running lights of type low beam, reduced low
beam or reduced high beam are constantly kept clean. On the other hand the other types of
running lights especially the low mounted special running lights often get very dirty.
They often have only half or less of the ori
ginal intensity. Another argument that also goes in favour of the main headlights is that since they are also used for night driving any faults (e.g. a burned out bulb) are quickly corrected.
In Finland there are proposals to make the limited daytime running light law general, that is to extend it over the whole year and
also include urban areas. In Norway the
autho-rities are discussing for and against a day time running light law. Denmark has a general daytime running light law for motorcycles and presently there is no intention to extend it to other vehicles. The main argument against a law
is of course the energy shortage.
The Swedish authorities are presently pre paring to make a suggestion to ECE (UN) con cerning an international reglementation for a day time running light. The idea is to coordinate the effort with the Finnish authorities which among other things probably would mean that the suggested acceptable intensity level will be somewhat higher than 300 800 cd.
The general conclusion is that daytime nmninglights by compensating the low visual peripheral sensitivity for oncoming vehicles, do function as an effective accident counter-measure.
REFERENCES
1. A. Burg, "Vision and Driving: A Report on Research." Human Factors 13, 1, 79 87, 1971
2. P.A. Davison, "The Role of Drivers' Vision in Road Safety." Light. Res. & Techn.
10, 3, 125 139, 1978.
3. H.R. Blackwell, "Contrast Thresholds of the Human Eye." Journal of the Optical Society
of America, 36, 624-643, 1946.
4. Nordic Traffic Safety Council, "Running Light Vehicle Lighting during Daylight." (In Swedish). Report 17, Liber, Stockholm, Sweden,
810191
5. K. Rumar, "Usage of Daytime Low Beam in Sweden 1967/68." (In Swedish). Unpublished re port from Dept. of Psychol., Univ. of Uppsala,
Sweden, 1968.
6. H. Sävenhed, "Daylight Use of Vehicle Lighting in Sweden during the Winter 1975/76."
(In Swedish). National Swedish Road and Traffic Research Institute, Linköping, Sweden, Meddel ande 37, 1977.
7. G. Von Bahr, "Is the Low Beam Law really Reasonable?" (In Swedish). Läkartidningen nr 19, 1971.
8. S. Dahlstedt & K. Rumar, Vehicle Colour and Front ConspiCuity in some Simulated Rural Traffic Situations." Unpublished report, Dept.
of Psychol., Univ. of Uppsala, Sweden, 1973.
9. U. Hörberg & K. Rumar, "Running Lights Conspicuity and Glare. Report 178, Dept. of Psychol., Univ. of Uppsala, Sweden, 1975.
10. U. Hörberg, "Running Light Twilight Conspicuity and Distance Judgement." Report 215, Dept. of Psychol., Univ. of Uppsala, Sweden, 1977.
11. U. Hörberg, "The Effect of Running Lights on Vehicle Conspicuity in Daylight and Twilight." Ergonomics, 22, 2, 165 173, 1979.
12. K. Rumar, "Running Lights Conspicuity, Glare and Accident Reduction." Accid. Anal. &
Prev., 12, 151 157, 1980.
13. L. Weström & L. Mårtensson, "Effects of Headlights on Detection of Vehicles by Rear View Mirrors." (In Swedish). Unpublished report from Dept. of Psychol., Univ. of Uppsala, Sweden, 1969.
14. O. Almkvist et al, "Estimation of
Vehicle Distance in Rear View Mirrors." (In
Swedish). Unpublished report from Dept. of Psychol., Univ. of Uppsala, Sweden, 1969.
15. L. Berlin & S. Björklund, "A Study of the Effect of Vehicle Colour on Distance Estima tion." (In Swedish). Unpublished report from Dept. of Psychc a, Univ. of Uppsala, Sweden, 1969.
16. B. Hisdal, "Evaluation of Accident Re ducing Effects by Running Lights at Daylight." (In Norwegian). Unpublished report from Sentral institutt for Industriell Forskning, Oslo,
Norway, 1975.
17. B. Färber et al, "Brake Lights Day-light Detectability with and without Lighted Tail Lights." Unpublished report from Dept. of Psychol., Univ. of Uppsala, Sweden, 1976.
18. B. Hisdal, "Running Lights." (In Nor wegain). Unpublished report from Sentralinsti-tutt for Industriell Forskning, Oslo, Norway, 1973.
19. Swedish Commission for Standardization.
Swedish Standard SS 3110: "Road Vehicles
Special Running Lights." SIS, Stockholm, Sweden, 1978.
20. L. Rindlöw, "Some Cost Calculations concerning Daytime Running Lights." (Personal communication), 1976.
21. H. Laurell & O. Odsell, "Possible Fuel
Savings with Cars of Today." (In Swedish).
National Swedish Road and Traffic Research
Institute, Linköping, Sweden, Report 157, 1978.
22. K. Andersson, G. Nilsson & M. Salus järvi, "Effects of Daytime Vehicle Lighting on
Road Accidents." (In Swedish). National Swedish
Road and Traffic Research Institute, Linköping,
Sweden, Report 102, 1976.
23. K. Andersson & G. Nilsson, "A Prelimi-nary Report on the Effect of Compulsory Daytime Running Lights on Accidents in Sweden." (In Swedish). Unpublished report from the Naticnal Swedish Road and Traffic Research Institute, Linköping, Sweden, 1979.
24. K. Andersson & G. Nilsson, "Effects of Daytime Running Lights on Accidents in Sweden after two Years." Preliminary data. (Personal Communication), 1981.
25. Swedish Road Safety Office, "The Effec tiveness of using Running Lights during Daytime in Sweden." Unpublished report from the Swedish Road Safety Office, Borlänge, Sweden, 1979.
26. B. Engdahl, "Drivers Using and not Using Daytime Low Beam. Some Characteristics and Attitudes." (In Swedish). Dept. of Communi-cation Technique, Technical University of
Stock-holm, Sweden, Report 1976:3, 1976.
27. Swedish Road Safety Office, "The Usage of Daytime Running Lights in August 1980." (In Swedish). Stat. grp. PM No. 29, Swedish Road Safety Office, Borlänge, Sweden, 1980.
This paper is subject to revision. Statements and For permission to publish this paper in full or in part, contact the
opinions advanced in papers or discussion are the SAE Publications Division.
author s and are his responsibility,not the Society s; Persons wishing to submit papers to be considered for presentation
however, the paper has been edited by SAE for uni- or publication through SAE should send the manuscript or a 300
form styling and format. Discussion will be printed word abstract of a proposed manuscript to: Secretary, Engineering
with the paper if it is published in SAE Transactions. Activity Board, SAE.