A.H.U. Bhatti et al.

Transportation Research Part D 99 (2021) 103025

Vehicle Dynamics Project Report

Development and analysis of electric vehicle driving cycle for

hill urban areas

Name Reg no
Muhammad Hassan Mustafa 313547

ME-41-C

Submitted to:
Dr. Raza Amer Azim

NUST COLLEGE OF
ELECTRICAL AND MECHANICAL ENGINEERING
PESHAWAR ROAD, RAWALPINDI

1
6
 A.H.U. Bhatti et al. Transportation Research Part D 99 (2021) 103025

Abstract

For the mountainous urban environment of Islamabad city, a real-world urban driving cycle with a
road slope profile is created in this work. A hybrid approach to data collecting is used, combining
digital surface models, on-board diagnostics, and global positioning systems. The Markov chain
Monte Carlo approach, which considered the weights of various road types in the research area,
was used to generate the driving cycle. With the help of 15 characteristics, 4 distribution
parameters, and the speed acceleration probability distribution, the generated driving cycle is
compared to other driving cycles. To determine key elements (i.e., energy consumption, range,
grid charging energy demand factors, and carbon dioxide emission factors), powertrain
simulations of 24 vehicle models of various vehicle technologies were run under 8 distinct driving
cycles. Results with validation show that, in addition to significant implications on the
components taken into consideration, the powertrain simulation would have had errors ranging
from 10.2 to 22.2% without the road slope profile.

1
7
 A.H.U. Bhatti et al. Transportation Research Part D 99 (2021) 103025

1. Introduction
One of the main reasons for the adaption of transport electrification is the rapid decline in urban
air quality and the depletion of fossil resources (Keramidas et al., 2020; Sun et al., 2017). Modern
vehicle technologies have enhanced energy recovery systems, efficient powertrain layout, and
transmission efficiency. In addition, the real-time energy usage display encourages better driving
habits, which results in lower energy consumption as compared to conventional internal
combustion engine vehicles (ICEVs) (Davari et al., 2017; Sayed et al., 2020). The issues of energy
consumption estimation, proper lifespan analysis, and public knowledge of the advantages of
electric vehicles (EV) must be overcome in order to assure a gradual transition to transport
electrification (Asghar et al., 2021; Ayodele and Mustapa, 2020; Capuder et al., 2020; P. Paniyil et
al., 2020; Sayed et al., 2020). A driving cycle (DC) is a speed profile that replicates the actual
traffic patterns in a certain area or city (Huertas et al., 2018). In addition to optimising battery
management, energy storage, and powertrain systems, DCs are used to design solutions to
problems. Additionally, DCs serve as a uniform testing process for assessing and approving
freshly manufactured automotive prototypes and automobiles (Brady and O'Mahony, 2016).
The federal test procedure, highway fuel economy driving schedule, new European driving cycle
(NEDC), urban dynamometer driving schedule (UDDS), Japanese industrial safety and health
association's 10-15 mode driving cycle, and JC 08 driving cycle are some of the common and
regulated DCs used globally for the various analyses of vehicles in different regions (Kurnia et al.,
2017; Yuhui et al., 2019; Zhao et al., 2020). These internationally regulated DCs differ from one
another in terms of data gathering region, data collection methodology, and DC development
methodology. Due to the variances in road conditions, geographic location, population density,
city planning, and traffic conditions, these DCs do not, however, correlate with actual driving
circumstances.

Fig. 1. Terrain of the administrative boundary of Islamabad.

1
8
 A.H.U. Bhatti et al. Transportation Research Part D 99 (2021) 103025

The worldwide harmonised light vehicles test method (WLTP) was created in 2015 to take the
position of the North European DC (NEDC) as a standard DC, to reduce analysis mistakes, and to
enhance the correlation of results with actual conditions (Tutuianu et al., 2013). The WLTP was
created using real-world driving and traffic data from five contributing nations, including the
USA, Europe, Japan, Korea, and India. As a result, the WLTP has a better correlation with actual
driving circumstances. A variety of global harmonised light vehicle test cycles (WLTC) are
included in the WLTP and are chosen for analysis of various cars based on their gravimetric power
density. These DCs are not appropriate in hilly urban terrains because none of them have a clear
road slope profile.
The analysis of the literature for several other research projects also shows that the energy
consumption and pollutant emissions computed using real-world DCs are higher than those
calculated using DCs that are governed by international law. The energy consumption of a plug-in
hybrid electric vehicle (PHEV) running under both NEDC and WLTP was presented by Liu et al.
The energy consumption of the PHEV in charge depletion (CD) mode was found to be 26%
greater than that of the NEDC. In contrast, the fuel consumption under WLTP is 20% greater than
NEDC when a PHEV is running in charge-sustaining (CS) mode (X. Liu et al., 2020). In the
Chinese city of Zhengzhou, researchers built a DC for hybrid electric buses using the Markov
chain Monte Carlo (MCMC) method, and they calculated the fuel consumption to be 20.5 L/100
km and the approximate energy consumption to be 12.8 kWh/100 km (Peng et al., 2020). It was
determined that the China light-duty vehicle test cycle results in a 13% increase in fuel
consumption by using traffic data from 40 different Chinese cities that were chosen based on their
GDP (Y. Liu et al., 2020). Another study team used the California performance measurement
system database's dataset to construct a global real-time DC creation process using the MCMC
method and dynamic programming. They found that the developed economic driving pro system
reduced fuel consumption by 5.8%. (Hongwen et al., 2018). None of them, however, has
conducted a composite examination of several vehicle technologies on numerous factors,
including energy consumption, driving range, simultaneous grid-required charging energy, and
road exhaust carbon dioxide (CO2) emissions.

1
9
 A.H.U. Bhatti et al. Transportation Research Part D 99 (2021) 103025

depicts the three-step process for creating DCs as recommended. The approach and framework for
analyzing the composite results of powertrain simulations for various vehicle technologies have
been carried out. For any worldwide metropolis with hilly urban environment, it can be used to
compile the viability of transport electrification and life cycle analysis. This work uses a DSM to
create a DC and create a road slope profile with the Islamabad driving cycle (IDC). Based on
proportional and characteristic parameters as well as speed acceleration probability distribution, an
extensive comparison of IDC with other DCs is made, including three classes of WLTC (SAPD).
Various vehicle technologies, including BEV, PHEV, HEV, FCEV, and ICEV, are simulated for
their powertrains across eight different DCs. The findings of the powertrain simulation are
compositely analyzed to determine energy consumption, range, necessary grid charge, and CO2
emissions.
This piece is divided into four sections. In Section 2, the selection of the test route, data collection
techniques, dataset preprocessing, and DC creation validation are discussed. In Section 3, the
outcomes of the generated DC's comparison to other common DCs are shown. Section 4 provides
examples of the powertrain simulation findings.

1
10
 A.H.U. Bhatti et al. Transportation Research Part D 99 (2021) 103025

2. Driving Cycle Creation

2.1. Test Route Determination

For the representative DC to be the true representative of urban driving conditions, a proper test
route must be selected which reflects all the topological structures of urban roads. The road
network dataset was obtained from open street maps (OpenStreetMap
Table 1
Road distribution of Islamabad district.
Road Type Color Length (km) Proportion (%) Mean Speed (km/h)

Primary Red 186 4.43 43.89

Secondary Blue 284 6.76 34.28
Tertiary Purple 730 17.39 26.88
Trunk Yellow 200 4.76 42.13
Service Orange 395 9.41 23
Residential Green 2404 57.25 27.26
Total – 4199 100 –

Fig. 3. a) Type of road in Islamabad, b) Selected test route data collection.

contributors, 2017). This dataset was processed in ArcGIS software to calculate the lengths of
different types of roads in Islamabad. The lengths, percentage proportions and mean speed of
different types of urban roads are shown in Table 1. Besides the topological dis- attribution of the
road network, other factors such as population density, different areas (residential, business,
recreational, commer- cial), and regional traffic conditions were considered for the determination
of the test route. The layout of the urban road network in Islamabad city and the selected test route
are illustrated in Fig. 3.

2.2. Data Collection

A GPS data logger with an Arduino nano ATmega328 microcontroller and a UBLOX UBX-
M8030-KT professional-grade chipset with navigation sensitivity of 167 dBm and a horizontal
sensitivity of 2-meter circle error probable was constructed to gather data (U-BloX, 2019). To
achieve high accuracy and data volume, the module was designed to solely use the GPS and the
data acquisition frequency was set to 1 Hz (Brady and O'Mahony, 2016). Toyota Aqua (Prius C)
2010 HEV was equipped with this GPS data logger, which was utilized to collect data.
1
11
 A.H.U. Bhatti et al. Transportation Research Part D 99 (2021) 103025

Date, Coordinated Universal Time, latitude, longitude, speed, heading, altitude, number of
satellites connected, and horizontal precision dilution are just a few of the parameters that the GPS
receiver keeps track of (HDOP). The GPS receiver must first be fixed for 26 seconds. After
starting the engine, the instrumented vehicle was maintained still for at least 30 seconds to provide
the GPS receiver enough time to connect with enough satellites to reduce the HDOP and record
correct data. True range multiliterate is the method used by the GPS receiver to determine position
and altitude. As contrast to horizontal position, the precision of altitude calculated by GPS
receivers is roughly halved. The GPS receiver is connected to the network of satellites above the
horizon to estimate the horizontal position. However, the lack of satellites below the horizon
makes it difficult to determine the altitude or vertical position, which drastically reduces vertical
position accuracy (Hofmann-Wellenhof et al., 2008).

Fig. 4. Relationship between key parameters of the collected dataset.

1
12
 A.H.U. Bhatti et al. Transportation Research Part D 99 (2021) 103025

2.3 Data Processing

2.3.1 Data Denoising and Smoothing
Due to many factors, including signal loss, tall structures, tree-lined roads, and driving conditions,
the data acquired by the sensors throughout the data collecting process tends to show noise and
distortion. Therefore, removing the noise component from the data signals is the first stage in the
data preparation procedure. The wavelet denoising method was utilized to effectively improve the
signal-to-noise ratio of the data signals by removing noise and distortion. The coefficients of
wavelet decomposition are generated for various frequency bands in the wavelet denoising
method. Then, without changing the coefficients of the input signal, the coefficients corresponding
to each frequency range are eliminated. Finally, to obtain a noise-free signal, wavelet
reconstruction is performed on the processed coefficients
2.3.2 Driving Fragment Togging
The dataset needs to be tagged with the proper driving elements in order to make the creation and
validation of the representative DC easier. Two levels of driving pieces are used to characterize
the driving process. Micro fragments and kinematic fragments make up these driving fragment
levels. The data fragment between two idle states is known as a micro fragment. As opposed to
this, kinematic fragments reflect operational states like acceleration, deceleration, idleness, and
cruising. By numerically differentiating the velocity data gathered by the GPS receiver, the
instantaneous acceleration ai is derived as follows:
ai = (vi+1 — vi/ti+1 — ti) (1000/3600), where I = 1, 2,.., k — 1. (1)
Owning to the factors such as accuracy of GPS receiver and road traffic conditions, the thresholds
for the division of kinematic fragments are set which are demonstrated in Table 2. A typical micro
fragment consists of all four kinematic fragments.
2.3.3 Markov chain Monte Carlo (MCMC) process
The representative DC was developed by using MCMC stochastic process. The MCMC process
is a probabilistic sampling technique in which the consecutive sample is dependent on the previous
sample. This dependency of samples is modeled by Markov chains for a
stochastic process possessing Markov property. A stochastic process fulfills the Markov property
if the state of the future event Xn+1
depends only on the current state Xn. This property can be expressed mathematically as:
P(Xn+1= x|X1 = x1, X2 = x2, ⋯, Xn = xn) = P(Xn+1= x|Xn = xn)
2) In the driving process, the future velocity is dependent on the current velocity of the
vehicle. Therefore, the driving process fulfills the Markov property and can be represented by
using Markov chains (Shi et al., 2016). To model the driving process using the Markov chain,
) is modeled as a process having ten states. The states of the driving
the driving process
process are defined using velocity ranges as shown in Table 3. The probability of single-step
transition from state xi to the state xj be expressed as:
pij = P Xn+1 = xj|Xn = xi
(3
1
13
 A.H.U. Bhatti et al. Transportation Research Part D 99 (2021) 103025

) As the driving process is modeled using ten states, the probability of transition from the current
state to any other future state can be represented using the transition probability matriX (TPM)
having dimensions of 10X10. The sum of each row of the TPM must be 1. The TPM is
computed using the following equation:
∑ ∑ )
pij = P Xn+1 = xj |Xn = xi = 1, ∀i

2.3.4 Validation of candidate driving cycle

= ⃒
N ⃒×

Given that the candidate DC is created using a stochastic process, it is essential

to use actual driving data to assess the candidate DC's correctness and representativeness. 20
typical parameters that characterize the driving dynamics were chosen to validate the proposed
DC. The candidate DC and actual driving data are used to compute these distinctive
characteristics. MAPE, which is determined by the relationship illustrated in equation (8), is used
to assess the discrepancy between the characteristic parameters of a candidate DC and actual data:
The IDC is created with a road slope profile as shown in Fig. 6. The characteristic parameters of the
IDC and the real-world driving data are given in Table 4. The comparison of the characteristic
parameters yielded MAPE of 5.9% only and the RMSE of SAPD is 3.4%. Hence, it is validated
that the constructed IDC is a good representative of real-world driving conditions.

1
14
 A.H.U. Bhatti et al. Transportation Research Part D 99 (2021) 103025

Fig. 6. a) IDC speed profile b) IDC road slope profile.

Table 4

Comparison of the representative driving cycle against the real-world dataset

Parameters Raw Data IDC Parameters Raw Data IDC

Idle Proportion (%) 15.897 15.333 95th Percentile of Acceleration (m/s2)
1.108 1.129
Acceleration Proportion (%) 37.129 38.083 RMS of Acceleration (m/s2) 0.523 0.541

Deceleration proportion (%) 25.627 25.250 Mean Deceleration 2 —0.487 —

2 (m/s )

0.511

Cruise Proportion (%) 21.347 21.333 SD of Deceleration (m/s ) 0.591 0.593

Mean Velocity (km/h) 29.509 29.533 5th Percentile of Deceleration

2 (m/s2) —
1.759 —1.803

SD of Velocity (km/h) 23.103 21.936 RMS of Deceleration (m/s ) 0.766 0.783

95th Percentile of Velocity (km/h) 72.095 69.952 Mean Slope (˚) 0.400
0.337

RMS of Velocity (km/h) 37.477 36.783 SD of Slope ˚

() 1.731 1.374

Mean Acceleration (m/s2) 0.385 0.392 95th Percentile of Slope ˚

() 3.084 2.376

SD of Acceleration (m/s2) 0.355 0.373 RMS of Slope (˚) 1.774 1.414

1
15
A.H.U. Bhatti et al. Transportation Research Part D 99 (2021) 103025

Table 5

Results of benchmark analysis.

Parameter NEDC UDDS LA92 US06 WLTC Class 1 WLTC Class 2 WLTC Class 3 IDC
Idle Proportion (%) 27.595 21.679 20.056 10.150 22.776 17.727 14.714
15.333

Acceleration Proportion (%) 21.435 28.394 35.306 31.281 20.235 28.755 28.762 38.083

Deceleration Proportion (%) 13.586 23.139 29.457 30.283 16.520 22.530 26.985 25.250

Cruise Proportion (%) 37.384 26.788 15.181 28.286 40.469 30.988 29.539 21.333

Maximum Velocity (km/h) 119.974 91.251 108.125 129.230 64.400 85.200 131.299 82.742

Mean Velocity (km/h) 33.202 31.508 39.595 77.197 28.472 35.718 46.499 29.533

SD of Velocity (km/h) 31.071 23.651 31.734 39.565 20.337 24.722 36.037 21.936

95th Percentile of Velocity (km/h) 99.978 82.166 99.436 116.838 58.779 73.715 119.000 69.952

RMS of Velocity (km/h) 45.464 39.391 50.736 86.730 34.984 43.434 58.823 36.783

Maximum Acceleration (m/s2) 1.055 1.475 3.084 3.755 0.805 1.000 1.750 2.411

Mean Acceleration (m/s2) 0.541 0.505 0.673 0.670 0.220 0.291 0.422 0.392

SD of Acceleration (m/s2) 0.206 0.451 0.514 0.816 0.163 0.216 0.379 0.373

95th Percentile of Acceleration (m/s2) 1.028 1.475 1.699 2.486 0.527 0.722 1.194
1.129

RMS of Acceleration (m/s2) 0.579 0.676 0.846 1.055 0.274 0.362 0.567 0.541

Maximum Deceleration (m/s2) —1.389 —1.475 —3.933 —3.084 —1.139 —1.167 —1.500 —2.889

Mean Deceleration (m/s2) —0.788 —0.578 —0.754 —0.728 —0.225 —0.335 —0.440 —0.511
2

SD of Deceleration (m/s ) 0.190 0.519 0.696 0.755 0.221 0.281 0.410 0.593

5th Percentile of Deceleration

2
(m/s2) —1.305 —1.475 —2.056 —2.298 —0.722 —0.868 —1.444
—1.803

RMS of Deceleration (m/s ) 0.811 0.776 1.025 1.048 0.316 0.437 0.601 0.783

MAPE of Parameters (%) 41.592 20.947 34.790 63.726 45.249 31.866 30.025 –

RMSE of SAPD (%) 10.990 7.140 8.528 23.133 15.258 8.263 10.159 –

3. Benchmark with other driving cycle

The internationally regulated DCs are compared to IDC to ascertain the variations in driving
situations. Three traditional urban DCs—NEDC, UDDS, and LA92—as well as US06, a
supplemental DC that simulates aggressive interstate driving, and three classes of WLTC test
cycles are contrasted with IDC for comparison's sake. The SAPD, characteristic parameters,
and kinematic proportion parameters are among the metrics used to compare these cycles.
Table 5 displays the outcomes of the benchmarking of these cycles. Figure 7 illustrates the

1
16
A.H.U. Bhatti et al. Transportation Research Part D 99 (2021) 103025

percentage proportions of velocity states and kinematic fragments, while Figure 8 depicts the
SAPD projections.
But when all the typical parameters were compared, the MAPE was 30.0% and the RMSE for
the SAPD was 10.2%. IDC has a lesser speed range but a wider range of acceleration and
deceleration, which explains why. Although UDDS has the lowest MAPE of 20.9% and
RMSE of SAPD of 7.1%, these errors are still too great for UDDS to be a reliable indicator of
driving habits in Islamabad.

4. Powertrain Simulation
Numerous techniques for powertrain simulations have been created since the pre-
commercialization phase of hybrid automobiles (between 1980 and 1990). Advisor, AVL Cruise
M, Autonomie, CarSim, HVEC, Simscape, V- EHLP, Dymola, SimulationX, OpenModelica,
JModelica, and the future automotive system technology simulator (FASTSim) are some of the
well-known tools (Tammi et al., 2016).

1
17
 A.H.U. Bhatti et al. Transportation Research Part D 99 (2021) 103025

Fig. 9. a) SOC of Chevrolet Bolt BEV under different driving cycles b) SOC of Chevrolet Bolt BEV under IDC with and without road slope.

Fig. 10. a) Speed profile of IDC, power required: b) Nissan leaf 30 kWh BEV, c) Toyota prius prime PHEV, d) Kia optima HEV, e) Toyota mirai FCEV,

f) Toyota corolla ICEV.

The FASTSim Python application programming interface (API) was chosen to be used for the
powertrain simulations in this work among all of these powertrain simulation tools. The reason for
this is that it uses a high-level simulation technique to mimic several powertrain configurations,
including BEV, PHEV, HEV, FCEV, and ICEV (Brooker et al., 2015). According to the
validation report for FASTSim, the RMSE for the fuel consumption simulation was only 5.6%.
(Gonder et al., 2018). The National Renewable Energy Laboratory made FASTSim's Python API
available for customization, therefore it was altered to meet the needs of this paper's calculation of
results. Table A1 provides the details of the vehicle models used for the research in this
publication. FASTSim simulates all 24 of the vehicles as they operate under 8 distinct DCs,
including IDC, UDDS, NEDC, LA92, US06, WLTC Class 1, WLTC Class 2, and WLTC Class 3.
Fig. 9 shows the state of charge (SOC) of the 2017 Chevrolet Bolt BEV operating under these DCs
(a). IDC and the road grade are both modelled for the powertrain simulations. The road grade has a
significant impact on the outcomes of any type of EV analysis and simulation.

1
18
 A.H.U. Bhatti et al. Transportation Research Part D 99 (2021) 103025

The SOC of the 2017 Chevrolet Bolt BEV, as simulated by the IDC with and without road slope,
is shown in Fig. 9(b). In hilly urban areas, different road slope profiles mean that an electric car's
battery pack must expend more energy to maintain the same speed profile. As a result, the SOC of
the battery pack decreases more quickly when the road slope profile is taken into account in the
simulation.
Fig. 10 shows the dynamic power results for various vehicles operating under the IDC, including
the Nissan Leaf 30 kWh BEV, Toyota Prius Prime PHEV, Kia Optima HEV, Toyota Mirai FCEV,
and Toyota Corolla ICEV. For the calculation of results for composite analysis, the DCs are
repeated until the fuel storage or in the case of BEV the energy storage gets depleted. The cluster
maps of simulated energy consumption in miles per gallon equivalent (MPGe) and vehicle range
are illustrated in Fig. 11 and Fig. 12 respectively.

1
19
 A.H.U. Bhatti et al. Transportation Research Part D 99 (2021) 103025

5. Conclusion
In this study, a hybrid data gathering methodology is used to generate an urban DC and a
road slope profile for steep urban terrain, such as Islamabad, Pakistan. The generated DC's
representation of actual driving situations was confirmed by an analysis of its various ratios and
defining characteristics. The IDC's road traffic conditions are closest to UDDS while still having
MAPE of 20.9% and RMSE of SAPD of 7.1%, according to a comparison of the IDC against four
internationally legislated DCs and three classes of WLTC. These DCs are not entirely realistic of
driving circumstances in the actual world, according to this comparison. To calculating the
equivalent energy consumption, range, grid charging need factors, and CO2 emission factors for 8
ICEVs, 6 BEVs, 5 PHEVs, 3 HEVs, and 2 FCEVs operating on the various DCs, powertrain
simulations were conducted for each of the powertrain types. The outcomes of the powertrain
simulation demonstrate that the road slope profile significantly affects EV. The findings of EVs
agglomerate RMSE of 10.2-13.5% in energy consumption, 10.2-13.4% in range, 11.5-12.2% in
grid charging energy required factors, and 17.6-22.2% in road CO2 emission factors if the road
slope is not considered. The comparison of the results of powertrain simulations using IDC and
other DCs revealed equivalent energy consumption error ranges from 6.2% to 34.7%, driving
range errors of 6.0% — 34.8%, grid charging energy requirement factors have errors of 3.6%
51.9%, and CO2 emission factor error ranges from 7.6% to 136.4%. To build the best energy and
battery management methods and to assess the effects of EVs running more accurately in
Islamabad on the electric grid, the generated DC may now be utilized for economic and life cycle
analysis. Additionally, the estimated distance-based gird charging requirements, road CO2
emission factors, and variable DCs are taken into consideration for realistic car models and
research projects that can use them in the future.
Future research will create an improved framework to address the gaps left by this study, which
will lead to even more precise assessments. The DSM utilized in this work has a resolution of 30
m, which is sufficient for hilly urban regions with broader roadways like Islamabad. This
restriction is the first. However, DSM of the higher resolution must be utilized in locations with a
higher percentage of small roads, which could raise the cost. The second restriction relates to the
powertrain simulations, where it is assumed that the study cars' auxiliary loads, such as lighting
and environment controllers, are constant as well as their passenger and luggage weights.
however, in the real world, these conditions vary from vehicle to vehicle and time to time.

1
20
A.H.U. Bhatti et al. Transportation Research Part D 99 (2021) 103025

References
6th Population & Housing Census, 2017 6th Population & Housing Census (Census), 2017. Pakistan Bureau of Statistics, Pakistan.

Asghar, R., Rehman, F., Ullah, Z., Qamar, A., Ullah, K., Iqbal, K., Aman, A., Nawaz, A.A., 2021. Electric vehicles and key
adaptation challenges and prospects in Pakistan: A comprehensive review. Journal of Cleaner Production 278, 123375.
https://doi.org/10.1016/j.jclepro.2020.123375.

Ayodele, B.V., Mustapa, S.I., 2020. Life Cycle Cost Assessment of Electric Vehicles: A Review and Bibliometric Analysis. Sustainability
12, 2387. https://doi.org/ 10.3390/su12062387.

Brady, J., O’Mahony, M., 2016. Development of a driving cycle to evaluate the energy economy of electric vehicles in urban ar eas.
Applied Energy 177, 165–178. https://doi.org/10.1016/j.apenergy.2016.05.094.

Brooker, A., Gonder, J., Wang, L., Wood, E., Lopp, S., Ramroth, L., 2015. FASTSim: A Model to Estimate Vehicle Efficiency, Cost and
Performance (SAE Technical Paper No. 2015- 01–0973). SAE International, Warrendale, PA. http://dx.doi.10.4271/2015-01-0973.

Capuder, T., Miloš Sprčić, D., Zoričić, D., Pandžić, H., 2020. Review of challenges and assessment of electric vehicles integration policy
goals: Integrated risk analysis approach. International Journal of Electrical Power & Energy Systems 119, 105894.
https://doi.org/10.1016/j.ijepes.2020.105894.

Davari, M.M., Jerrelind, J., Stensson Trigell, A., 2017. Energy efficiency analyses of a vehicle in modal and transient driving cycles
including longitudinal and vertical dynamics. Transportation Research Part D: Transport and Environment 53, 263–275.
https://doi.org/10.1016/j.trd.2017.04.019.

Demuynck, J., Bosteels, D., De Paepe, M., Favre, C., May, J., Verhelst, S., 2012. Recommendations for the new WLTP cycle based on an
analysis of vehicle emission measurements on NEDC and CADC. Energy Policy, Special Section: Fuel Poverty Comes of Age:
Commemorating 21 Years of Research and Policy 49, 234–242. https://doi.org/10.1016/j.enpol.2012.05.081.

Galgamuwa, U., Perera, L., Bandara, S., 2016. Development of a driving cycle for Colombo, Sri Lanka: an economical approach for
developing countries. Journal of Advanced Transportation 50 (7), 1520–1530. https://doi.org/10.1002/atr.v50.710.1002/atr.1414.

Gonder, J.D., Brooker, A.D., Wood, E.W., Moniot, M., 2018. Future Automotive Systems Technology Simulator (FASTSim) Validation
Report No. NREL/TP–5400- 71168, 1462683). https://doi.org/10.2172/1462683.

Günther, R., Wenzel, T., Wegner, M., Rettig, R., 2017. Big data driven dynamic driving cycle development for busses in urban public
transportation. Transportation Research Part D: Transport and Environment 51, 276–289.
https://doi.org/10.1016/j.trd.2017.01.009.

He, H., Guo, J., Zhou, N., Sun, C., Peng, J., 2017. Freeway Driving Cycle Construction Based on Real-Time Traffic Information and
Global Optimal Energy Management for Plug-In Hybrid Electric Vehicles. Energies 10, 1796.
https://doi.org/10.3390/en10111796.

Ho, S.-H., Wong, Y.-D., Chang, V.-W.-C., 2014. Developing Singapore Driving Cycle for passenger cars to estimate fuel consumption
and vehicular emissions.

Atmospheric Environment 97, 353–362. https://doi.org/10.1016/j.atmosenv.2014.08.042.

Hofmann-Wellenhof, B., Lichtenegger, H., Wasle, E., 2008. GNSS – Global Navigation Satellite Systems: GPS, GLONASS, Galileo,
and more. Springer-Verlag, Wien. https://doi.org/10.1007/978-3-211-73017-1.

Hongwen, H., Jinquan, G., Jiankun, P., Huachun, T., Chao, S., 2018. Real-time global driving cycle construction and the application to
economy driving pro system in plug-in hybrid electric vehicles. Energy 152, 95–107.
https://doi.org/10.1016/j.energy.2018.03.061.

Huertas, J.I., Díaz, J., Cordero, D., Cedillo, K., 2018. A new methodology to determine typical driving cycles for the design of vehicles
power trains. Int J Interact Des Manuf 12 (1), 319–326. https://doi.org/10.1007/s12008-017-0379-y.

Hung, W.T., Tong, H.Y., Lee, C.P., Ha, K., Pao, L.Y., 2007. Development of a practical driving cycle construction methodology: A
case study in Hong Kong.

Transportation Research Part D: Transport and Environment 12 (2), 115–128. https://doi.org/10.1016/j.trd.2007.01.002.

1
21
A.H.U. Bhatti et al. Transportation Research Part D 99 (2021) 103025

Jing, Z., Wang, G., Zhang, S., Qiu, C., 2017. Building Tianjin driving cycle based on linear discriminant analysis. Transportation
Research Part D: Transport and Environment 53, 78–87. https://doi.org/10.1016/j.trd.2017.04.005.

Keramidas, K., Tchung-Ming, S., Diaz-Vazquez, A.R., Weitzel, M., Vandyck, T., Tamba, M., Krause, J., Van Dingenen, R., Chai, Q.,
Fu, S., Wen, X., Soria-Ramirez, A., European Commission, Joint Research Centre, 2020. Global energy and climate outlook 2019:
electrification for the low-carbon transition : the role of electrification in low-carbon pathways, with a global and regional focus on
EU and China.

Kurnia, J.C., Sasmito, A.P., Shamim, T., 2017. Performance evaluation of a PEM fuel cell stack with variable inlet flows under
simulated driving cycle conditions.

Applied Energy 206, 751–764. https://doi.org/10.1016/j.apenergy.2017.08.224.

Liu, X., Zhao, F., Hao, H., Chen, K., Liu, Z., Babiker, H., Amer, A.A., 2020a. From NEDC to WLTP: Effect on the Energy
Consumption, NEV Credits, and Subsidies Policies of PHEV in the Chinese Market. Sustainability 12, 5747.
https://doi.org/10.3390/su12145747.

Liu, Y.u., Wu, Z.X., Zhou, H., Zheng, H., Yu, N., An, X.P., Li, J.Y., Li, M.L., 2020b. Development of China Light-Duty Vehicle Test
Cycle. Int.J Automot. Technol. 21 (5), 1233–1246. https://doi.org/10.1007/s12239-020-0117-5.

Ntziachristos, L., Samaras, Z., 2020. EMEP/EEA air pollutant emission inventory guidebook 2019 (No. 1 .A.3.b.i), Passenger Cars.
European Monitoring and Evaluation Programme/European Enviroment Agency.

OpenStreetMap contributors, 2017. Planet dump retrieved from https://planet.osm.org.

Paniyil, P., Singh, R., Powar, V., Venayagamoorthy, G.K., 2020. In: Sustainable Power for Electrification of Transportation. 2020
Clemson University, in, pp. 1–7. https://doi.org/10.1109/PSC50246.2020.9131313.

Pavlovic, J., Ciuffo, B., Fontaras, G., Valverde, V., Marotta, A., 2018. How much difference in type-approval CO2 emissions from
passenger cars in Europe can be expected from changing to the new test procedure (NEDC vs. WLTP)? Transportation
Research Part A: Policy and Practice 111, 136–147. https://doi.org/ 10.1016/j.tra.2018.02.002.

Peng, J., Jiang, J., Ding, F., Tan, H., 2020. Development of Driving Cycle Construction for Hybrid Electric Bus: A Case Study in
Zhengzhou. China. Sustainability 12, 7188. https://doi.org/10.3390/su12177188.

Pouresmaeili, M.A., Aghayan, I., Taghizadeh, S.A., 2018. Development of Mashhad driving cycle for passenger car to model vehicle
exhaust emissions calibrated using on-board measurements. Sustainable Cities and Society 36, 12–20.
https://doi.org/10.1016/j.scs.2017.09.034.

Quirama, L.F., Giraldo, M., Huertas, J.I., Jaller, M., 2020. Driving cycles that reproduce driving patterns, energy consumptions and tailpipe
emissions. Transportation Research Part D: Transport and Environment 82, 102294. https://doi.org/10.1016/j.trd.2020.102294.

Sayed, K., Kassem, A., Saleeb, H., Alghamdi, A.S., Abo-Khalil, A.G., 2020. Energy-Saving of Battery Electric Vehicle Powertrain and
Efficiency Improvement during Different Standard Driving Cycles. Sustainability 12, 10466. https://doi.org/10.3390/su122410466.

Shi, S., Lin, N., Zhang, Y., Cheng, J., Huang, C., Liu, L., Lu, B., 2016. Research on Markov property analysis of driving cycles
and its application. Transportation Research Part D: Transport and Environment 47, 171–181.
https://doi.org/10.1016/j.trd.2016.05.013.

Suhaimi, M.F.M., Salleh, N.A.M., Sahari, D.T.M.M., Sarip, M.S., 2020. Development of Kuala Lumpur driving cycle for the estimation
of fuel consumption and vehicular emission. IOP Conf. Ser.: Mater. Sci. Eng. 834, 012040 https://doi.org/10.1088/1757-
899X/834/1/012040.

Sun, L., Huang, Y., Liu, S., Chen, Y., Yao, L., Kashyap, A., 2017. A completive survey study on the feasibility and adaptation of EVs in
Beijing, China. Applied Energy 187, 128–139. https://doi.org/10.1016/j.apenergy.2016.11.027.

Sun, Z., Wen, Z., Zhao, X., Yang, Y., Li, S., 2020. Real-World Driving Cycles Adaptability of Electric Vehicles. World Electric Vehicle
Journal 11, 19. https://doi.org/ 10.3390/wevj11010019.

1
22
 A.H.U. Bhatti et al. Transportation Research Part D 99 (2021) 103025
Tammi, K., Minav, T., Kortelainen, J., 2018. Thirty Years of Electro-Hybrid Powertrain Simulation. IEEE Access 6, 35250–35259. https://doi.org/10.1109/
Access.628763910.1109/ACCESS.2018.2850916.

Tutuianu, M., Marotta, A., Heinz, S., Eva, E., Takahiro, H., Noriyuki, I., Hajime, I., 2013. Development of a World-wide Worldwide harmonized Light duty driving Test Cycle (WLTC) (No.
GRPE-68-03), UN/ECE/WP.29/GRPE/WLTP-IG.

U-BloX, 2019. UBX-M8030-CT/KT/KA Product Summary.

Yuhui, P., Yuan, Z., Huibao, Y., 2019. Development of a representative driving cycle for urban buses based on the K-means cluster method. Cluster Comput 22 (S3),6871–6880.
https://doi.org/10.1007/s10586-017-1673-y.

Zhang, J., Wang, Z., Liu, P., Zhang, Z., Li, X., Qu, C., 2019. Driving cycles construction for electric vehicles considering road environment: A case study in Beijing.

Applied Energy 253, 113514. https://doi.org/10.1016/j.apenergy.2019.113514.

Zhao, X., Ma, J., Wang, S., Ye, Y., Wu, Y., Yu, M., 2019. Developing an electric vehicle urban driving cycle to study differences in energy consumption. Environ Sci Pollut Res 26
(14), 13839–13853. https://doi.org/10.1007/s11356-018-3541-6.

Zhao, X., Zhao, X., Yu, Q., Ye, Y., Yu, M., 2020. Development of a representative urban driving cycle construction methodology for electric vehicles: A case study inXi’an.
Transportation Research Part D: Transport and Environment 81, 102279. https://doi.org/10.1016/j.trd.2020.102279

1
23
A.H.U. Bhatti et al. Transportation Research Part D 99 (2021) 103025

1
24