Roberto De Lieto Vollaro et.

al / International Journal of Engineering and Technology (IJET)

Energy Performance Optimization of a Bus

for Urban Public Transport
Roberto De Lieto Vollaro#, Fabio Botta #1, Luca Evangelisti#, Paola Gori#, Claudia Guattari#
#
Department of Engineering, Università degli studi Roma Tre, Via della Vasca Navale 79, Rome - 00146, Italy
1
fabio.botta@uniroma3.it
Abstract— Nowadays many people use public transports in urban centres. Consequently, every day a
lot of buses move within cities, trying to ensure the best service to citizens. During the year buses become
crowded places and using an air conditioning system in constant operation, it tries to ensure a certain
condition of comfort for the passengers on board. The aim of this study is to analyse and optimize the
energy performance of a bus shell, identifying practical solutions that have not yet been adopted in order
to reduce the impact of air conditioning on bus consumption and, therefore, on air pollution. For this
reason it was decided to conduct a thermal analysis of a bus for public transport, in order to understand
the behaviour of the bus shell and to deduce possible optimization measures that have not yet been made
until now. The analysis was carried out considering the hottest day of July and the coldest day of
January, considering the conditions of operation based on the most common graphics TGM able to define
the concentration of traffic city during the 24 hours per day. The study was done using the dynamic
simulation software TRNSYS. With this software it was possible to recreate faithfully the structure of the
bus and the external environmental conditions, assessing the impact of different technical solutions for an
improvement of internal conditions and a reduction of the cooling capacity required. Because the
presence of passengers in public transport is considered like a “benefit” during the winter, the analysis
started with the identification of a summer solution and the subsequent evaluation of this solution for the
wintertime. The aim of this study was to optimize the bus shell and select the most appropriate solutions.
Regarding the transparent surfaces it has been given importance to factors such as the thermal
transmittance and the solar gain factor (g-value). Aware of the influence given by the solar radiation on
the energy loads, we have simulated bus energy performance also considering different types of
paintwork with high reflectance.
I. INTRODUCTION
Nowadays a lot of people spend many hours during the day using public transport. Traffic conditions in
Rome necessarily require the frequent use of public buses. In Italy conditioning the public transport arises from
the need to improve the comfort of people who daily use public services moving around the city. Internal
temperature and humidity conditions are important factors in terms of passengers comfort and health. For this
reason, the automobile industry has recently developed the thermal comfort issue which have to take care of
both the health of the passengers and the driver. The problem connected with the comfort onboard is very
complex because, inside, there are strong gradients of air velocity and temperature. Moreover the solar radiation
inside the vehicle, especially during the summer, is very important [1]. In this study it has been developed a
model applying dynamic simulation software TRNSYS (already largely used to analyze the thermal exchanges
[2]-[5]). In order to give a better understanding of energy behavior of the bus shell, the identification of practical
solutions that have not been used yet, the improvement of the isolation from the outside and the limitation of the
power required for cooling the cabin have been carried out.
II. MODELLING
The analysis was carried out considering a generic model of a bus and its geometrical characteristics are
shown in Fig. 1. The vehicle stratigraphy is given in Table II and III. To provide the thermal analysis the
dynamic simulation software TRNSYS has been used. Through this programme it was possible to faithfully
reproduce both the geometry of the bus and the external environmental conditions: the bus modelling is done
through the TRNSYS Build, which is the section of the software that allows to generate the vehicle’s model
only. The connection of the "build" with the outside world is possible by using the TRNSYS Studio, through
which it is possible to connect specific types, generating a pattern of interconnection (Fig. 2).
A. Trnsys Build
Geometry of the vehicle (Fig. 1) and stratigraphy have been inserted through TRNSYS Build. The materials
used into the model are shown in Table II and III. Considering the initial configuration of the vehicle (called
“standard configuration”) it was used a single glass and its characteristics are reported in Table IV. The solar
absorption coefficient of the blue paint, commonly adopted for buses, was considered to be equal to 0.6 [6].
Moreover, the model was oriented to have the larger surfaces exposed to the south and north respectively, and it

ISSN : 0975-4024 Vol 5 No 4 Aug-Sep 2013 3639

 Roberto De Lieto Vollaro et.al / International Journal of Engineering and Technology (IJET)

was considered a variable number of passengers on board, Table I [7]. Having knowledge that TRNSYS is
always connected with the energy performance of buildings, it was generated a model which is the nearest to the
operating conditions of a city bus. With this aim, they were appropriately used convective coefficients that could
simulate the relative motion of the air in the model, in order to simulate the bus movement [8]. In this way it
was possible to vary the convective coefficients considering the bus motion [9].

Fig. 1. Bus model

TABLE I
Daily Passengers on Board

Step time Passengers on board Step time Passengers on board

6:00 20 15:00 40
7:00 80 16:00 40
8:00 80 17:00 80
9:00 40 18:00 80
10:00 40 19:00 40
11:00 40 20:00 40
12:00 40 21:00 40
13:00 80 22:00 20
14:00 40 23:00 1
TABLE II
Characteristics of Vertical Opaque Walls

Thickness Density Conductivity [kJ/h m Heat Capacity [kJ/kg

Material Resistivity [hm2K/kJ]
[mm] [kg/m3] K] K]
Fiberglas
8 2500 0.7920 1.256
s
Air 0.034
Fiberglas
8 2500 0.7920 1.256
s

TABLE III
Characteristics of Horizontal Opaque Walls

Thickness Density Conductivity [kJ/h m Heat Capacity [kJ/kg Resistivity

Material
[mm] [kg/m3] K] K] [hm2K/kJ]
Fiberglass 8 2500 0.7920 1.256
Wood
20 350 0.36 1.88
panel
Air 0.034
Fiberglass 8 2500 0.7920 1.256

TABLE IV
Characteristics of Transparent Walls

Material Thickness [mm] Transmittance [W/m2 K] g-value

Glass 8 5.68 0.855

ISSN : 0975-4024 Vol 5 No 4 Aug-Sep 2013 3640

 Roberto De Lieto Vollaro et.al / International Journal of Engineering and Technology (IJET)

B. Trnsys Studio
The model connection with the outside world is made through the TRNSYS Studio. Fig. 2 allows to see how the
software uses external data-sheet (Type RH/T), entering values of relative humidity and ambient temperature. In
this way the direct use of weather-data - that contains airport meteorological-data - hasn’t been applied and this
because it is not always correct to simulate the external weather conditions. Despite this, the weather-data (Type
109-TMY2) were still considered to use information relating to solar radiation. Type33e receives as input
relative humidity values and ambient temperature values to calculate the dew point temperature. Type69b
calculates the fictive sky temperature. Calculators, represented in Fig. 2, allow to perform mathematical
operations between data contained in different types. Type56a - representing the model and using the generated
one through TRNSYS Build –

Fig. 2. TRNSYS Studio piping

allows to create a connection with the "outside world." Finally, Type65d graphically displays the thermal loads
and Type25a allows to get a spread sheet with data related to loads, calculated hour by hour. Referring to
critical conditions, the analysis was performed by reproducing the environmental conditions in summer and
winter: more in depth, it was considered July 13th and January 11th. The choice to use a dynamic software, such
as TRNSYS, to simulate the energy performance of a bus shell is connected with the high reliability and
effectiveness of the software analysis. The higher level of detail achievable is related to the possibility of taking
into account the realistic temperature fluctuations and, therefore, a more realistic calculation of the energy loads.
The steps of the analysis are shown in Fig. 3.

Fig. 3. Analysis scheme with TRNSYS

ISSN : 0975-4024 Vol 5 No 4 Aug-Sep 2013 3641

 Roberto De Lieto Vollaro et.al / International Journal of Engineering and Technology (IJET)

III. BUS SHELL EFFICIENCY EVALUATION

An air conditioning system (hot / cold) - characterized by infinite power and set-point temperatures of 27°C
for the summer and 20°C for the winter - has been considered to evaluate the shell efficiency.
Starting from the analysis of this configuration, simulations have been performed according to the following
steps:
Summer
 Intervention on transparent surfaces
 Intervention on opaque surfaces
 Intervention on colours
 SUMMER OPTIMAL SOLUTION
Winter
 Checking the summer optimal solution
The first step is based on the energy behaviour analysis, improving the technical characteristics of the glass.
The characteristics that we focused on are:
• the thermal transmittance value;
• the value of the solar gain factor (g-value).
Following the first step, which is not change the stratigraphy of the opaque walls and acts on transparent
surfaces only, it was analysed the vehicle’s behaviour by adopting for the summer period (July 13th) windows
shown in Table V.
TABLE V
Technical Characteristics of Used Glasses

Material Thickness [mm] Transmittance [W/m2 K] g-value

Glass type 2 Double glazing 4/16/4 2.83 0.755
Glass type 3 Float Glass 19 5.16 0.682
Glass type 4 Double glazing 4/16/4 0.98 0.440
Obtained the best results, it was performed the second type of intervention changing only the opaque walls. In
this second step, it was interposed between the fiberglass panels a polyurethane foam, its characteristics are
shown in Table VI. Different simulations were performed - considering 1 cm, 2 cm and 3 cm of foam – waiting
for a power request reduction.
TABLE VI
Technical Features of the Thermal Insulation

Material Density [kg/m3] Conductivity [kJ/h m K] Heat Capacity [kJ/kg K]

Polyurethane Foam 33 0.023 1.674
Using the foam insulation, it was obtained a reduction of the opaque walls thermal transmittance as reported
in Table VII. Regarding the horizontal walls, 1 cm of foam has been incorporated into stratigraphy, getting a
thermal transmittance value of 1.208 W/m2K, which has been compared to 1.941 W/m2K related to standard
configuration.

ISSN : 0975-4024 Vol 5 No 4 Aug-Sep 2013 3642

 Roberto De Lieto Vollaro et.al / International Journal of Engineering and Technology (IJET)

Fig. 4 - Solar radiation on surfaces (winter)

Fig. 5 - Solar radiation on surfaces (summer)

TABLE VII
Insulation Thickness and Walls Overall Transmittance

Polyurethane Foam (vertical walls) Insulation thickness [mm] Entire wall Transmittance [W/m2 K]
Standard_wall 0 2.739
Wall_10 10 1.476
Wall_20 20 0.899
Wall_30 30 0.646

Before the comparison between different solutions, it was analyzed the solar radiation and ambient
temperature along the 18 hours per day (both summer and winter). In this way it was possible to understand
which is the most critical time interval. The results are shown in Figs. 4, 5 and Figs. 6 and 7.

ISSN : 0975-4024 Vol 5 No 4 Aug-Sep 2013 3643

 Roberto De Lieto Vollaro et.al / International Journal of Engineering and Technology (IJET)

External temperature, July 13th

60.00

Temperature [°C]
40.00
20.00
TEMP ext
0.00
1.00 6.00 11.00 16.00 21.00
Time

Fig. 6 - External temperature, July 13th

Outside temperature, January 11th

Temperature [°C]

15.00
10.00
5.00 Temp EXT
0.00
0.00 5.00 10.00 15.00 20.00
Time

Fig. 7 - External temperature, January 11th

IV. SUMMER RESULTS

Starting from the standard configuration, there were applied the following steps:
 Intervention on transparent surfaces
 Intervention on opaque surfaces
 Intervention on colours
 SUMMER OPTIMAL SOLUTION
A. Intervention on Transparent Surfaces
Referring to the first step, it has been performed the comparison between the bus standard configuration
(standard glass) and the types of glasses shown in Table V, adopting a value of solar absorbance of 0.6 for
opaque surfaces. Adopting glasses with high performance, it has been obtained the results shown in Fig. 8, that
show the daily service thermal load (sum of thermal loads calculated hour by hour).

Energy loads as a function of type of glass

100.00

95.00
Thermal Load

glass type 2;
95.52
[kWh]

90.00 standard
glass; 93.01 glass type 3;
91.74
85.00 glass type 4;
87.62
80.00

Fig. 8 – Energy loads as a function of type of glass

Table VIII shows the differences between standard configuration of the bus and the different type of glasses
used (referring to a single work-day).

ISSN : 0975-4024 Vol 5 No 4 Aug-Sep 2013 3644

 Roberto De Lieto Vollaro et.al / International Journal of Engineering and Technology (IJET)

TABLE VIII
Percentage Difference Between Thermal Loads

Glass Type Thermal load [kWh] Percentage difference [%]

Standard glass 93.01 -
Glass type 2 95.52 2.70
Glass type 3 91.74 -1.36
Glass type 4 87.62 -5.79

B. Intervention on Opaque Surfaces

It was made the comparison between the bus standard configuration (stratigraphy of opaque walls shown in
Table IV) and the new configuration in which the polyurethane foam (as a thermal insulator) has been
considered. The simulations have been performed using 1 cm, 2 cm and 3 cm of foam within the opaque walls.
Fig. 9 shows the results carried out.

Thermal loads variation in function of foam layer

thickness
100.00
Foam
98.00
Thermal Load [kWh]

Foam 3cm; 99.25

2cm; 98.38
96.00 Foam
1cm; 96.68
94.00
92.00 standard
wall; 93.01
90.00
88.00

Fig. 9 – Thermal loads variation as a function of foam layer thickness

Fig. 9 points out that, using the foam, there is an increase of thermal loads. Therefore, it is clear that this
intervention cannot be considered suitable to optimize the energy performance of the bus shell.
C. Intervention on Colours
The last phase of the analysis involves the use of paints characterized by a reduced solar absorption
factor (light colours). Thermal loads were calculated by considering at the beginning classic blue paint colour
(characterized by a solar absorbance of 0.6) and then a white paint (characterized by solar absorbance equal to
0.14). The colour’s effects on thermal loads are shown in Fig. 10. There is a difference between the two
solutions and it is equal to 10.65%.

Thermal load as a function of color absorbance

95.00
Thermal Load [kWh]

90.00 Blue; 93.01

85.00

80.00
White; 83.10
75.00

Fig. 10 - Thermal load as a function of colour absorbance

The Fig. 10 shows the difference between conventional paints and white paints. A thermal loads
reduction equal to 10.65% is associated to the use of a high-reflective paint.
V. DISCUSSION
In order to optimize its energy performance and to reduce the energy request in summer, some
interventions on the bus shell have been done. The changes are:
• the use of high performance glasses, compared to conventional single glasses (thickness of 4 mm);

ISSN : 0975-4024 Vol 5 No 4 Aug-Sep 2013 3645

 Roberto De Lieto Vollaro et.al / International Journal of Engineering and Technology (IJET)

• the use of an insulating foam;

• the use of a paint with high reflectance (white paint).
Using the Glass type 2 - double glaze characterized by a lower thermal transmittance value - there is a thermal
load increase. This is connected with the lower value of the thermal transmittance of the glass and this produces
a heat accumulation inside the bus shell. These considerations can be better understood by considering the basic
equation that defines the cooling load [5].
Qcool = (Qsol + Q int) − ηu (Qdisp + Qvent ) (1)

In the Eq. (1) the terms related to the internal gains and ventilation are constants. Using Glass type 2
there is a variation connected only with the heat losses. The solar gains are unchanged and the total cooling load
reach higher values.
Using the Glass type 3 - single glass characterized by a high thermal transmittance and a g-value
smaller than the standard glass it is possible to assess a load reduction. The importance of the g-value is due to
the meaning of the solar gain factor, which can be defined as the sum of two parameters: an optical component
and a thermal component. The optical component takes the name from the optical transmittance while the
thermal component from the emissivity of the inner side (proportional to the optical-absorbance). The solar
radiation between 300 nm and 2500 nm absorbed by the external side of the window is released through the
internal side with a longer wavelengths (medium infrared and far infrared), in the form of heat. Referring to the
equation (1), which expresses the cooling load, it is possible to assess that using Glass type 3 a reduction of the
solar gains, connected with the lower g-value, and the thermal dispersion, Qdisp, that is the same to the standard
configuration, has been recorded.
Using the Glass type 4 - glass that has the lowest thermal transmittance and the lowest g-value – it was
obtained a thermal heat load lower than the standard configuration’s value. Using a glass with these
characteristics, the decrease energy demand can be correlated to a simultaneous reduction of the solar gains
(Qsol) and of the shell dispersion capacity (Qdisp) . In this case the transmittance is more effective than the solar
gain factor.
The effect of insulating foam, placed between the opaque surfaces, represents an incorrect choice to
optimize the shell’s performances during summer. Referring, once again, to the Eq.(1), it is possible to assert
that the presence of a thermal insulator leads to a lower thermal flows dispersion with a consequent increase of
the load itself.
The last intervention is characterized by the application of a high reflective paint. The white colour has
a solar absorbance of 0.14 and it allows a greater solar radiation reflection compared to the conventional paints
today used.
VI. SIMULTANEOUS EFFECTS
Analysing the simultaneous benefits using glasses and colours with high features, it was performed a
simulation using both the glass type 4 and the white paint. It is clear why we haven’t taken in account the use of
the insulating foam.
The results, compared with the standard configuration of the bus, are shown in Fig. 11.

Maximum efficiency of the bus shell

95.00

90.00
Thermal Load [kWh]

Standard
85.00 Config.; 93.01

80.00

75.00 Optimized
Config. ; 78.28
70.00

Fig. 11 – Benefits associated with the simultaneous use of high performance glazing and high-reflectance paint

The thermal load reduction compared with the starting configuration of the vehicle is equal to 15.84%.
Because of a bus use during the whole year, it was also performed a simulation in winter conditions, considering
the optimized shell (white paint and glass type 4). The results obtained during the coldest day of January
(January 11th) and compared with the standard configuration of the vehicle, are shown in Fig. 12 and Fig. 13.

ISSN : 0975-4024 Vol 5 No 4 Aug-Sep 2013 3646

 Roberto De Lieto Vollaro et.al / International Journal of Engineering and Technology (IJET)

Thermal loads for standard and optimized bus

configurations
6.00

Thermal load [kWh]

Optimized Bus Standard Bus
3.00

0.00
6.00 11.00 16.00 21.00
time
Fig. 12 - Thermal loads for standard and optimized bus configurations

Inside temperatures for standard and optimized

bus configurations
30.00
Optimized inside Temp. Standard inside Temp.
Inside Temperatures [°C]

27.00

24.00

21.00

18.00
6.00 11.00 16.00 21.00
Time

Fig. 13 - Inside temperatures for standard and optimized bus configurations

Figs. 12 and 13 show as the optimization procedure implies a reduction of the thermal load, while the
internal temperature remains substantially unchanged in the two cases. The air conditioning system is used only
during the last hours of the daily service, however, it is possible to assess a load reduction.
Considering the bus engine power - from 245 HP to 290 HP [10] - and a bus weight of 12080 kg, it was
analysed the power/weight ratio concerning the use of the glass type 4. Using a double-glazing instead of a
conventional glass means an increase of vehicle’s weight. The specific weight of the glass and the weight/power
ratio are shown in Table IX. Using the glass type 4, there is an increase of the bus weight of approximately 310
kg. For this reason the total weight of the bus is equal to 12390 kg, Table IX, and it was used to calculate the
weight/power ratio for an engine with 245 hp and 290 hp.
TABLE IX
Analysis of the Weight / Power Ratio

Single weight Transparent surface Total weight

Standard config. bus 12080 kg
Glass type 4 20 kg/m2 31 m2 620 kg
Standard glass 10 kg/m2 31 m2 310 kg
Bus with standard glass 12080 kg
Bus with glass type 4 12390 kg

Engine power 245 CV 290 CV

Standard config. bus 49.3 kg/CV 41.65 kg/CV
Optimized config. bus 50.6 kg/CV 42.72 kg/CV
Engine efficiency 0.35
Alternator efficiency 0.9
Air conditioning efficiency 3
Diesel cost 1.67 €/liter [11]

ISSN : 0975-4024 Vol 5 No 4 Aug-Sep 2013 3647

 Roberto De Lieto Vollaro et.al / International Journal of Engineering and Technology (IJET)

Using data shown in Table IX, a fuel annual saving of 20% has been estimated. Considering the fleet of
vehicles which is owned by a public transportation company, it is possible to reach high economic savings.
It is important to observe that an annual saving - in terms of liters of fuel - can involve a reduction of CO2
emissions into the environment. Being aware of the influence of CO2 on the greenhouse effect, optimizing the
bus with the proposed solutions and lowering the fuel consumption, the reduction of CO2 was estimated. This
value is approximately equal to 1464 kg/year per bus.
VII. PRELIMINARY ANALYSIS OF THE CONSTRUCTION COST
A cost analysis associated to the use of glass type 4 has been joined to the energetic analysis. Considering a
transparent surface of 31 m2 and an estimated price of glass of 140.00 €/m2, the additional cost for each bus is
4340.00 Euros. Estimating a standard bus price of 150000.00 Euros, it is possible to state that using glass type 4
a surplus of the starting amount of about 3% has been recorded (Table X). This analysis should be considered as
a preliminary one because it was not taken into account the relationship between the bus producer and the
manufacturer of insulating glass units. The relationship between bus manufacturers and glasses producer should
lead to a cost reduction of vehicle’s production. In this costs analysis was not considered the use of high
reflectance paint because it is already included into the global bus cost.
TABLE X
Efficiency Costs Analysis

UNIT COST TRANSPARENT SURFACE GLOBAL COST

STANDARD CONFIG. BUS 150000,00 € 150000.00 €
GLASS TYPE 4 140.00 €/M2 31 m2 4340.00 €
TOTAL 154340.00 €
SURPLUS ≅3 %
VIII. CONCLUSIONS
Today many people use public transport and it is important to improve the bus comfort, especially in critical
conditions such as rush hours and summer time. For this reason it is important to improve energy performances
of a bus shell, under critical conditions, such as summer weather (high temperatures) and maximum number of
passengers on board. There were evaluated, through a dynamic simulator, several interventions, such as using
more efficient glasses and paints characterized by a low solar absorbance value. In order to reduce the thermal
loads, it is possible to improve the energy performances of vehicles adopting a glass characterized by a thermal
transmittance value equal to 0.98 W/m2K and a solar gain factor equal to 0.44 (specifications glass type 4).
Moreover, they have been calculated the thermal loads by adopting a high reflectance paint (white color) and the
benefit of both solutions - in terms of energy demand - equal to 15.84% has been evaluated.
The shell optimization allows to adopt interventions that can be considered suitable also during the winter.
Aiming to achieve the global efficiency of the bus shell, it was estimated the cost of a conventional bus. Then
it was evaluated the cost of a glass (per square meter) having the same performances of the glass type 4 used in
the simulations. The preliminary costs analysis leads to a price’s increase of about 3%, but it brings, at the same
time, to an annual fuel saving of 20% for the diesel vehicle.
REFERENCES
[1] R. De Lieto Vollaro, S. Grignaffini, A. Vallati, “Numerical analysis and measures for the evaluation of comfort inside buses used for
public transport,” in Proc. Urban Transport XIV: Urban Transport and the Environment in the 21st Century, 2008, Malta, vol. 101, pp.
373-383.
[2] TRNSYS, User manual TRNSYS 16.
[3] M. S. Al-Homoud, “Computer-aided building energy analysis techniques,” Building and Environment, vol. 36, pp. 421–433, 2001.
[4] S. A. Kalogirou, G. Florides, S. Tassou, “Energy analysis of buildings employing thermal mass in Cyprus,” Renewable Energy, vol. 27,
pp. 353–368, 2002.
[5] UNI TS 11300 Determinazione del fabbisogno di energia termica dell’edificio per la climatizzazione estiva ed invernale.
[6] Regolamento n. 36 sulle prescrizioni uniformi relative alle caratteristiche costruttive degli autobus aventi una capacità di oltre 16
posti a sedere o in piedi, escluso il conducente – Commissione Economica per l’Europa delle Nazioni Unite.
[7] Piano strategico per la mobilità sostenibile, Linee di indirizzo – Comune di Roma, 2009.
[8] D. L. Loveday, A. H. Taki, “Convective heat transfer coefficients at a plane surface on a full-scale building façade,” International
Journal of Heat and Mass Transfer, vol. 39, pp. 1729–1742, 1996.
[9] UNI EN ISO 6946 “Componenti ed elementi per edilizia, resistenza termica e trasmittanza termica – metodo di calcolo”.
[10] Irisbus Iveco Citelis, autobus standard 3 porte, diesel.
[11] Ministero dello Sviluppo Economico - Prezzi Medi Nazionali Mensili, 2013.

ISSN : 0975-4024 Vol 5 No 4 Aug-Sep 2013 3648