Transportation Research Part D 17 (2012) 509–513

Contents lists available at SciVerse ScienceDirect

Transportation Research Part D

journal homepage: www.elsevier.com/locate/trd

Influence of vehicle driving parameters on the noise caused

by passenger cars in urban traffic
J.A. Calvo a,⇑, C. Álvarez-Caldas a, J.L. San Román a, P. Cobo b
a
Departamento de Ingeniería Mecánica, Universidad Carlos III de Madrid, Avda. De la Universidad, n°30, 28911 Leganés, Spain
b
Centro de Acústica Aplicada y Evaluación No Destructiva, CSIC-UPM, Serrano 144, 28006 Madrid, Spain

a r t i c l e i n f o a b s t r a c t

Keywords: In this work, a sample of vehicles has been instrumented to measure of variables that influ-
Vehicle noise emissions ence vehicle noise emissions in Madrid. A circuit reproducing a normal travel pattern in
Urban traffic noise large city is traveled by a fleet of vehicle models representing the fleets of cars in a Euro-
Aggressive driving pean city. A sample of drivers covers the test track under different traffic conditions. Driv-
Driving behavior
ing parameters and noise emitted have been recorded in each test and average values have
been extracted. These data have been analyzed to define the noise emissions produced by a
vehicle in real driving conditions and to identify the noisiest driving behaviors.
Ó 2012 Elsevier Ltd. All rights reserved.

1. Introduction

About 20% of the EU population are estimated to suffer noise levels considered to be unacceptable, and another 45% to live
in areas where noise can cause serious annoyance (Affenzeller and Rust, 2005). On average, about half of the noise in urban
areas is produced by road traffic, and this has become a cause of social tension between economic development and quality
of life in some places. Despite the efforts made to reduce the noise emissions of individual vehicles, noise disturbance is not
diminishing largely because of the growth of the vehicle fleet.
Current abatement legislation in Europe is based on the noise level of a vehicle according to test procedures in Directive
70/157/EEC (European Economic Community, 1970). If a vehicle exceeds the maximum level established, it is approved for
use. The maximum noise level depends on the category and type of the vehicle and has been declining gradually with up-
dates of the Directive. Given traffic growth, however, his reduction has done little to reduce the overall traffic noise (Sand-
berg, 2001). One reasons for this is that the test does not properly represent the noise level emitted by a car in real driving
conditions but is rather based on specified, specific conditions.
To determine noise emission levels in real driving conditions, we initially establish the parameters that cab influence on
noise emission (Ochieng et al., 2004) and then take a representative sample of vehicles and drivers to represent average driv-
ing behavior based on a route that approaches average driving condition in large cities (Ericsson, 2000).
The drivers behave as they usually do with one exception, an expert driver who drives aggressively to allow comparison
of such behavior with normal driving parameters and noise emissions. This allows identifying of driving behavior that has
particularly adverse effects on vehicle noise and parameters that could be used to detect this.

2. Methodology

Passenger car vehicles representative of the current Spanish fleet are studied for their noise contribution (Asociación
Nacional de Fabricantes de Automóviles y Camiones, 2009). The following vehicles are examined: B-segment vehicles (com-

⇑ Corresponding author. Tel.: +34 916248791.

E-mail address: jacalvo@ing.uc3m.es (J.A. Calvo).

1361-9209/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.trd.2012.06.002
510 J.A. Calvo et al. / Transportation Research Part D 17 (2012) 509–513

Table 1
Characteristics of selected vehicles.

Vehicle Segment Engine type Engine size (cm3) Vehicle mass (kg) Tire size
Seat Ibiza B Diesel 1598 1170 185/60 R14
Seat Ibiza B Gasoline 1390 1049 185/60 R15
VW Golf C Diesel 1968 1314 205/55 R16
Opel Astra C Gasoline 1598 1373 205/55 R16
Audi A4 D Diesel 1968 1475 215/55 R16
Mercedes C180 D Gasoline 1796 1485 225/40 R18

pact cars) that amounted to 30% of the Spanish fleet in 2010; C-segment vehicles (midsize cars), 29% of the fleet; and D-seg-
ment vehicles, 15% of the fleet. Other segments have a much smaller impact on urban traffic and are no included.
For each segment, a diesel engine model and a gasoline engine model have been selected. Brands and models have been
chosen after analyzing the vehicle sales registration statistics. Table 1 summarizes their characteristics:
The test path has been selected according to:

 Representativeness: According to (Banister et al., 2000) over 75% of the EU population live in urban areas and around 20%
of all miles traveled are urban trips. The circuit is thus representative of the average urban journey that a driver takes in
his daily activities (commuting, shopping, etc.) These types of trips account for about 80% of trips in large cities. The aver-
age trip on urban roads is between 8 and 12 km.
 Traffic density: Traffic levels in the selected area are high but not outside of the range of many cities.
 Noise emissions level: The selected area, taken from the Strategic Noise Map of the Madrid City (Ayuntamiento de Madrid,
2006), embraces zones afflicted with high traffic noise levels.

Carabanchel, the circuit selected in Madrid, is divided into zone according to noise nuisance (Fig. 1) based on day equiv-
alent continuous noise levels (Fig. 1). The test track has a day equivalent continuous noise level of between 70 and 75 dB (A)
and traffic a density of 20,000–40,000 vehicles per day. It includes roads with a variety of speed limits, generally 50 km/h, but
some sections are limited to 30 km/h. Half of it involves two-lane two-ways streets, and the other 50% runs through four-
lanes, two-ways streets. The circuit is 8500 m long and has 25 traffic lights and three roundabouts. Tests are carried out using
the different drivers randomly. Drivers perform the circuit taking into account traffic conditions and the tests are carried out
different times (morning and afternoon) and on several days of the week. They involved 21 drivers, selected to cover a range
of driving skills (Jackson et al., 2006);

 Five men, with more than 5 years of driving experience (Driver category A).
 Five men, with less than a year of driving experience (Driver category B).
 Five women, with more than 5 years of driving experience (Driver category C).
 Five women, with less than 1 year of driving experience (Driver category D).
 One professional driver (to simulate aggressive driving) (Driver category E).

The professional driver deliberately drove aggressively to try to reduce travel time. He drove fast and selected gears lower
than normal to increase the acceleration of the vehicle when making continuous lane changes. The rest of the drivers drove
in their same usual way taking into account traffic conditions.
In terms of technology, a microphone was placed inside the engine space, near the inlet valve, and a second close to the
rear wheel, opposite the exhaust pipe (Robertson et al., 1998).

3. Results

A vehicle’s speed is recorded when it is in motion along with stopped time to calculate its average speed. The difference in
average speeds of drivers by category is only around 7% compared to the aggressive driver that is around 20%. No significant
differences is found in the average speed for vehicles by driver category, including the ‘‘aggressive’’ driver, indicating vehicle
types does not greatly influence average speed when driving. Taking all drivers and vehicles, and assuming they represent a
normal distribution, with driver E excluded from the calculation, the average value of the ‘‘normal’’ drivers is 29.7 km/h1
whereas it is 37.7 km/h for the aggressive driver.
The novice drivers, categories C and D, are 7% slower when driving larger vehicles than more experienced drivers, and
men with more experience (category A) are significantly faster than other categories across all vehicles. On the other hand,
the variation of speed is 30% less for the standard drivers than for the aggressive one reflecting the smoother driving. Stan-
dard drivers use second and third gears most, with the aggressive driver preferring first and second.

1
To average 29.7 km/h it is necessary to exceed the speed limit at least 17% of the time.
 J.A. Calvo et al. / Transportation Research Part D 17 (2012) 509–513 511

Fig. 1. Day equivalent continuous noise levels in the ‘‘Carabanchel’’ district and the trial circuit.

The time a vehicle is stopped is similar in all cases and it depends mainly on traffic conditions, traffic lights, etc. The
aggressive driver spends slightly less time shifting gears. Both values appear independent of the driver and vehicle types
in the context of standard drivers. Engine speeds are analyzed in the same way. The average value for standard drivers is
1800 rpm with a standard deviation of 575 rpm that matches the zone of maximum torque and, thus, minimum fuel con-
sumption. In the case of the aggressive driver, the average value is 2750 rpm and the standard deviation 1026 rpm. The stan-
dard driver’s behavior leads to lower engine speeds and smooth driving, compared to aggressive driver.
To analyze the influence of the driving factors on noise, Leq,1s graphs and levels histogram from the microphones are used.
Because drivers A–D exhibit similar driving characteristics, we use them as the ‘‘standard driver’’ and compare them with the
‘‘aggressive driver’’. The upper chart of Fig. 2 shows the Leq,1s results for diesel and the bottom for gasoline engines. The
sound level clearly falls when a vehicle is stopped and increases when it is accelerating. The engine noise is also higher
for diesel than for gasoline engines, mainly at lower values when the vehicles are idling. Fig. 3 shows the trend in Leq,1s
for the wheel microphone for standard and aggressive drivers. As expected, wheel noise is highly correlated with the speed
of each vehicle, and thus, the wheel noise emitted by the aggressive driver is significantly higher than that by the standard
driver. No significant differences are found between fuel types.
Table 2 shows the overall Leq for the two types of driver. In the case of gasoline engines, the aggressive driver is 9 dB nois-
ier than the standard one in terms of engine, and 4.7 dB noisier in terms of wheel, but they are similar in case of standard
drivers. In the case of the diesel, engine noise for the aggressive driver is 7.4 dB and wheel noise is 4 dB noisier than the aver-
age over the other categories. As in the case of the gasoline engine, wheel is similar to the engine noise for the standard driv-
ers. The aggressive driver, however, produces more engine than wheel noise (4.1 dB).

Fig. 2. Leq,1s graph of vehicle engine noise – diesel (top); gasoline (bottom).
512 J.A. Calvo et al. / Transportation Research Part D 17 (2012) 509–513

Fig. 3. Leq,1s graph of the vehicle wheel noise – diesel (top); gasoline (bottom).

Table 2
Global equivalent levels Leq for engine and wheel noises for every driver category.

Driver category Gasoline engine Diesel engine

Engine noise (dB) Wheel noise (dB) Engine noise (dB) Wheel noise (dB)
A 103.2 104.5 107.9 106.8
B 102.6 103.5 106.9 106.9
C 102.2 102.6 106.8 105.9
D 102.3 102.5 106.6 105.6
Average (A–D) 102.6 103.4 107.1 106.3
E 111.7 108.1 114.5 110.4

4. Conclusions

The average behavior of a sample of vehicles and drivers is analyzed in real urban traffic conditions to produce an average
behavior for a ‘‘standard driver’’ and an ‘‘aggressive driver’’. Neither driver experience nor sex has any appreciable impact on
noise generated, and engine type has little effect. Engine noise is significantly greater, however, for aggressive driving, and
especially so for gasoline vehicles because they tend to be able to go faster and accelerate more rapidly, but under analogous
conditions, engine noise is always greater for diesel vehicles than for gasoline. At low and medium engine speeds, gasoline
vehicles also produce less noise than diesel ones.

Acknowledgments

This work has been supported by the Spanish Ministry of Science and Innovation (MICINN) through Grants Nos. TRA2008-
05654-C03-02 and TRA2008-05654-C03-03.

References

Affenzeller, J., Rust, P., 2005. Road Traffic Noise – A Topic for Today and the Future. VDA Technical Congress, Ingolstadt.
Asociación Nacional de Fabricantes de Automóviles y Camiones, 2009. Matriculaciones turismos y vehículos comerciales, Madrid.
Ayuntamiento de Madrid, 2006. Mapa Estratégico de Ruido de la Ciudad de Madrid. Madrid.
Banister, D., Stead, D., Sheen, P., Akerman, J., Dreborg, K., Nijkamp, P., Schleicher-Tappeser, 2000. European Transport Policy and Sustainable Mobility. Spon,
London.
Ericsson, E., 2000. Variability in urban driving patterns. Transportation Research Part D 5, 337–354.
European Economic Community, 1970. Council Directive 70/157/EEC of 6 February 1970 on the Approximation of the Laws of the Member States Relating to
the Permissible Sound Level and the Exhaust System of Motor Vehicles. Brussels.
Jackson, E., Qu, Y., Holmén, B.A., Aultman-Hall, Lisa, 2006. Driver and Road Type Effects on Light-duty Gas and Particulate Emissions. Transportation
Research Board, 85th Annual Meeting, Washington DC.
 J.A. Calvo et al. / Transportation Research Part D 17 (2012) 509–513 513

Ochieng, W.Y., Quddus, M.A., Noland, R.B., 2004. Integrated Positioning Algorithms for Transport Telematics Applications. Institute of Navigation Annual
Conference, Long Beach.
Robertson, S. et al, 1998. The Effect of Speed on Noise, Vibration and Emissions from Vehicles. VTT Communities & Infrastructure, Finland.
Sandberg, U., 2001. Noise Emissions of Road Vehicles. Effect of Regulations. Final Report 01–1, I-INCE WP-NERV, Stockholm.