Optimization data on total cost of ownership for conventional and battery electric heavy vehicles driven by humans and by automated driving systems

In road freight transport, the emerging technologies such as automated driving systems improve the mobility, productivity and fuel efficiency. However, the improved efficiency is not enough to meet environmental goals due to growing demands of transportation. Combining automated driving systems and electrified propulsion can substantially improve the road freight transport efficiency. However, the high cost of the battery electric heavy vehicles is a barrier hindering their adoption by the transportation companies. Automated driving systems, requiring no human driver on-board, make the battery electric heavy vehicles competitive to their conventional counterparts in a wider range of transportation tasks and use cases compared to the vehicles with human drivers. The presented data identify transportation tasks where the battery electric heavy vehicles driven by humans or by automated driving systems have lower cost of ownership than their conventional counterparts. The data were produced by optimizing the vehicle propulsion system together with the loading/unloading schemes and charging powers, with the objective of minimizing the total cost of ownership on 3072 different transportation scenarios, according to research article “Impact of automated driving systems on road freight transport and electrified propulsion of heavy vehicles” (Ghandriz et al., 2020) [2]. The data help understanding the effects of traveled distance, road hilliness and vehicle size on the total cost of ownership of the vehicles with different propulsion and driving systems. Data also include sensitivity tests on the uncertain parameters.


a b s t r a c t
In road freight transport, the emerging technologies such as automated driving systems improve the mobility, productivity and fuel efficiency. However, the improved efficiency is not enough to meet environmental goals due to growing demands of transportation. Combining automated driving systems and electrified propulsion can substantially improve the road freight transport efficiency. However, the high cost of the battery electric heavy vehicles is a barrier hindering their adoption by the transportation companies. Automated driving systems, requiring no human driver on-board, make the battery electric heavy vehicles competitive to their conventional counterparts in a wider range of transportation tasks and use cases compared to the vehicles with human drivers. The presented data identify transportation tasks where the battery electric heavy vehicles driven by humans or by automated driving systems have lower cost of ownership than their conventional counterparts. The data were produced by optimizing the vehicle propulsion system together with the loading/unloading schemes and charging powers, with the objective of minimizing the total cost of ownership on 3072 different transportation scenarios, according to research article "Impact of automated driving systems on road freight transport and electrified propulsion of heavy vehicles" (Ghandriz et al., 2020) [2]. The data help understanding the effects of traveled distance, road hilliness and vehicle size on the total cost of ownership of the vehicles with different propulsion and driving systems. Data also include sensitivity tests on the uncertain parameters.
© 2020 The Author(s). Published by Elsevier Inc. This is an open access article under the CC BY license.
( http://creativecommons.org/licenses/by/4.0/ ) Specifications Table   Subject Engineering Specific subject area Transportation, Energy, Automotive Engineering and mathematical optimization Type of data Chart and Graph How data were acquired By implementing mathematical simulation and optimization using models of roads and vehicle dynamics Data format Raw and Analyzed Parameters for data collection Roads of different lengths and hillinesses and vehicles of different size and driving system, i.e., human-driven or driven by automated driving systems.

Description of data collection
The optimum Heavy vehicle propulsion system and corresponding infrastructure with the lowest total cost of ownership was found in terms of size of the internal combustion engine, the type and number of battery packs, type and number of electric motors, charging powers from charging stations and loading/unloading (LU) scheme. The optimization was repeated for roads of different lengths and hillinesses and vehicles of different size and driving system, i.e., human-driven or driven by automated driving systems.

Value of the data
• These data provide the total cost of ownership and cost components of deploying different heavy vehicles in the various transportation scenarios and help to perform a comparative assessment between different road freight transport solutions. • These data provide valuable information for the practitioners, vehicle manufactures and transportation companies with regard to competitiveness of the battery electric heavy vehicles and automated driving systems (i.e., high and full driving automation) against the conventional heavy vehicles and vehicles driven by human drivers. • These data help further research in adoption of the automated driving systems and battery electric heavy vehicles in road freight transport. • These data identify those transportation scenarios, where the battery electric heavy vehicles become competitive to the conventional combustion-powered heavy vehicles. • Data provide sensitivity of the total cost of ownership to the different parameters such as the utilization level and fuel efficiency covering a wide range of transportation scenarios in the different geographical regions. Table 1 Content of file "mainDataOptTCO.mat" Variable name Description adsBEHV _ MissionTimes * A 4-dimensional cell array containing the mission times of the automated driving system-dedicated battery electric heavy vehicles (ADS BEHVs) adsCHV _ optPropInfra * A 4-dimensional cell array containing the propulsion hardware, loading-unloading and charging powers, explained in Table 2 , belonging to the automated driving system-dedicated conventional combustion-powered heavy vehicles (ADS CHVs) hdBEHV _ TCOcomponents * A 4-dimensional cell array containing the components of TCO belonging to human-driven battery electric heavy vehicles (HD BEHVs) hdCHV _ TCOPerYearPerTon * A 4-dimensional cell array of the optimum TCO belonging to human-driven conventional combustion-powered heavy vehicles (HD CHVs) * A similar variable exists for the other vehicles and driving systems.

Data description
Three data files were provided.
1. mainDataOptTCO.mat, that includes the total cost of ownership (TCO), cost components of TCO, duration time of a round-trip including charging and LU, and the optimum setup of vehicle propulsion and infrastructure of all the 3072 transportation scenarios. The content of the file and the variable names are explained in Table 1 . Each variable is a 4-dimensional cell array corresponding to the road hilliness, road length, vehicle size and average trip speed. The data were produced assuming 100% utilization rate (i.e. vehicle-time on operation), and using the nominal values of vehicles and cost parameters provided in [2] , when there was a single vehicle in the fleet. 2. sensitivityData.mat, that includes the sensitivity tests of TCO and its cost components with respect to the fuel price, vehicle utilization, battery price, fuel efficiency, economic life time, discount rate, automation-specific hardware price and electric energy price. Sensitivity tests were done for optimum average speed and for all roads, vehicles and driving systems, i.e., human-driven or driven by the automated driving systems, when there was a single vehicle in the fleet. 3. roadData.mat, that includes topographic data, i.e., elevation and road grade versus distance of all road types.
The data were analyzed and related figures were provided in this article. Furthermore, Matlab codes used for generating the figures from the raw data were provided in files "opti-mumTCOgraphs.m" and "sensitivityFigures.m". File "optimumTCOgraphs.m" also provides the optimum propulsion system and infrastructure of all vehicles, that drive in an optimum average speed, as an output, given a road length and hilliness. Moreover, the number of vehicles of the same type in the fleet can be selected.

Experimental design, materials, and methods
A transportation scenario is defined by the following parameters.
• Vehicle size, that can be one of the followings: rigid truck (RT), tractor-semitrailer (TS), Nordic combination (NC), and A-double (AD), with the gross mass of 25 ton, 40 ton, 60 ton, and 80 ton, respectively. • Driving system, i.e., human-driven (HD) or driven by the automated driving systems with no human driver on-board. According to [1] J3016, the latter is called automated driving systems-dedicated vehicle (ADS-DV).
The combination of all the parameter choices above resulted in the 3072 different transportation scenarios. For generating the data, on each of the transportation scenarios, the following optimization problem was solved to optimize the vehicle propulsion system and infrastructure including the LU scheme and charging powers, with the objective of minimizing the TCO. The design variables and their discrete ranges are according to Table 2 .
As an objective function, the TCO included the operational costs and depreciation of purchase cost. The purchase cost included the cost of chassis, driver cabin, transmission, automated driving systems hardware, ICE, battery packs, EMs, LU, charging stations, and the investment on transportation mission management system (TMMS) needed for managing ADS-DVs. The operational costs included the cost of driver, diesel fuel, electric energy, maintenance, tax, insurance, and the operational costs related to TMMS. In calculating the TCO, the details such as the driver rest time and battery degradation and replacement were considered. Moreover, the difference in the rest value of the battery and other vehicle hardware was taken into account, depending on the battery state of health when vehicle service life ends. Furthermore, the optimization constraints ensure the vehicle proper operation on each of the transportation scenarios using models of vehicle dynamics. More details were provided in [2] .
Finally, after solving the optimization problems, the sensitivity tests were performed on all the roads and for all the vehicles but only for an average reference speed that yielded the lowest TCO. Fig. 1 depicts the process of data generation and presentation. The process of data generation and presentation. Please refer to [2] for the nominal values of the vehicles specification and cost parameters. The selected 8 parameters used for sensitivity tests include the fuel price, vehicle utilization, battery price, fuel efficiency, economic life time, discount rate, automation-specific hardware price and electric energy price..

Particle swarm optimization algorithm
The defined optimization problem (1) is nonlinear and non-smooth. The stochastic optimization methods and in particular the particle swarm optimization (PSO) showed to be effective methods for solving such problems, according to [3] , [5] and [4] ; however, the global optimum may not be found unless solving the optimization problem is repeated many times. The presented data include the best attained solution among 20 optimization runs which took about 12 days to run on a computer with 32 cores and 92 GB of RAM.
PSO stochastically and iteratively moves a population of particles, i.e., points in the search space, closer to an optimum solution. PSO uses the speed and direction of particles' motion relative to their own positions or the position of the best particle that is found in the previous iteration. PSO algorithm according to [5] is described as follows.
1. Determine the range of design variables a k in Table 2 , i.e. the search space, and their minimum and maximum values; or alternatively their indices k , since the variables may belong to discrete and/or non-numeric sets. In that case, the design variables must be arranged in an increasing order if applicable.
Initialize the swarm, i.e., the position of particles k and their speeds v , randomly, and the cost function C t .
where · is the closest integer function, r s,ij and r v, ij denote random numbers in [0,1], C t denotes total cost of ownership per unit freight, k p i is the previous position of the i th particle, k gp denotes previous global best position, N denotes the number of particles in the swarm, and n denotes the number of dimensions of the search space, or the number of design variables ( n = 9 ).
3. Evaluate the cost function C t (a k i ) , for i = 1 , . . . , N, according to [2] . 4. Update the best global position and position of each particle, for i = 1 , . . . , N, if its current position is better than previous position. k 5. Update velocities and positions, for i = 1 , . . . , where, q and r denote random numbers in [0,1], t = 1 is the time step, and w is a number (usually in [0,1]) that determines the influence of the previous velocity on the current velocity. 6. Return to step 3 if termination criteria is not satisfied.

Analyzed data
The data can be used for choosing the right vehicle in terms of the vehicle size, propulsion system, and the driving system (i.e., human-driven or driven by the automated driving systems), knowing the properties of the transportation scenario such as the road length and road hilliness. Provided figures handle multidimensionality of the data and facilitate data interpretation and comparative analysis. Tables 3-6 explain the content of figures. The figures presented in Tables 3-5 , were produced assuming 100% utilization rate (i.e. vehicle-time on operation), and using the nominal values of vehicles specification and cost parameters provided in [2] .           Table 6 The sensitivity tests of the optimized TCO and its components to the different parameters are shown for the different vehicles and driving systems, and for the optimum speed of the transportation scenario. The vehicle size, road hilliness and road length are fixed in a figure, but they vary between figures according to this In figures, the chassis cost includes also the cost of driver cabin, transmission and automated driving systems hardware.
The TCO optimizations were done while there was a single vehicle in the fleet. However, file "optimumTCOgraphs.m" provides the possibility to regenerate the data (other than sensitivity data) for more number of vehicles than one, with a fixed propulsion system and infrastructure, assuming that the optimum propulsion system and infrastructure are not influenced by the number of fleet vehicles. However, for the sensitivity figures, if the number of vehicles in a transportation scenario is more that one, the cost components relating to charging station (and possibly LU) must be divided by the number of vehicles. Furthermore, if the cost of the charging infrastructure is intended to be excluded from the TCO, for example, if it is provided by a third party, then it must be disregarded in the figures.
Refer to [2] for better interpretation of data presented in the figures.  Table 3 for the other vehicle sizes and road hillinesses.  Table 3 for the other vehicle sizes and road hillinesses.  Table 3 for the other vehicle sizes and road hillinesses.  Table 3 for the other vehicle sizes and road hillinesses.  Table 3 for the other vehicle sizes and road hillinesses.  Table 3 for the other vehicle sizes and road hillinesses.  Table 3 for the other vehicle sizes and road hillinesses.  Table 3 for the other vehicle sizes and road hillinesses.  Table 3 for the other vehicle sizes and road hillinesses.  Table 3 for the other vehicle sizes and road hillinesses.  Table 3 for the other vehicle sizes and road hillinesses.  Table 3 for the other vehicle sizes and road hillinesses.  Table 3 for the other vehicle sizes and road hillinesses.  Table 3 for the other vehicle sizes and road hillinesses.  Table 3 for the other vehicle sizes and road hillinesses.  Table 3 for the other vehicle sizes and road hillinesses.  Table 4 for the other road types.  Table 4 for the other road types.  Table 4 for the other road types.  Table 4 for the other road types.  Table 4 for the other road types.  Table 4 for the other road types.  Table 4 for the other road types.  Table 4 for the other road types.  Table 4 for the other road types.  Table 4 for the other road types.  Table 4 for the other road types.  Table 4 for the other road types.  Table 4 for the other road types.  Table 4 for the other road types.  Table 4 for the other road types.  Table 4 for the other road types.  Table 4 for the other road types.  Table 4 for the other road types.  Table 4 for the other road types.  Table 4 for the other road types.  Table 4 for the other road types.  Table 4 for the other road types.  Table 4 for the other road types.  Table 4 for the other road types.  Table 5 for the other road types.  Table 5 for the other road types.  Table 5 for the other road types.  Table 5 for the other road types.  Table 5 for the other road types.  Table 5 for the other road types.  Table 5 for the other road types.  Table 5 for the other road types.  Table 5 for the other road types.  Table 5 for the other road types.  Table 5 for the other road types.  Table 5 for the other road types.  Table 5 for the other road types.  Table 5 for the other road types.  Table 5 for the other road types.  Table 5 for the other road types.  Table 5 for the other road types.  Table 5 for the other road types.  Table 5 for the other road types.  Table 5 for the other road types.  Table 5 for the other road types.  Table 5 for the other road types.  Table 5 for the other road types.  Table 5 for the other road types.  Table 6 for the other vehicle sizes and road types.  Table 6 for the other vehicle sizes and road types.  Table 6 for the other vehicle sizes and road types.  Table 6 for the other vehicle sizes and road types. driving systems are shown for the optimum speed of the transportation scenario. A group of four bars from left to right represent BEHV HD, BEHV ADS-V, CHV HD and CHV ADS-DV, respectively. See Table 6 for the other vehicle sizes and road types.  Table 6 for the other vehicle sizes and road types.  Table 6 for the other vehicle sizes and road types.  Table 6 for the other vehicle sizes and road types.  Table 6 for the other vehicle sizes and road types.  Table 6 for the other vehicle sizes and road types.  Table 6 for the other vehicle sizes and road types.  Table 6 for the other vehicle sizes and road types.  Table 6 for the other vehicle sizes and road types.  Table 6 for the other vehicle sizes and road types.  Table 6 for the other vehicle sizes and road types.  Table 6 for the other vehicle sizes and road types.  Table 6 for the other vehicle sizes and road types.  Table 6 for the other vehicle sizes and road types.  Table 6 for the other vehicle sizes and road types.  Table 6 for the other vehicle sizes and road types.  Table 6 for the other vehicle sizes and road types.  Table 6 for the other vehicle sizes and road types.  Table 6 for the other vehicle sizes and road types.  Table 6 for the other vehicle sizes and road types.  Table 6 for the other vehicle sizes and road types.  Table 6 for the other vehicle sizes and road types.  Table 6 for the other vehicle sizes and road types.  Table 6 for the other vehicle sizes and road types.  Table 6 for the other vehicle sizes and road types.  Table 6 for the other vehicle sizes and road types.  Table 6 for the other vehicle sizes and road types.  Table 6 for the other vehicle sizes and road types.  Table 6 for the other vehicle sizes and road types.  Table 6 for the other vehicle sizes and road types.  Table 6 for the other vehicle sizes and road types.  Table 6 for the other vehicle sizes and road types.  Table 6 for the other vehicle sizes and road types.  Table 6 for the other vehicle sizes and road types.  Table 6 for the other vehicle sizes and road types.  Table 6 for the other vehicle sizes and road types.  Table 6 for the other vehicle sizes and road types.  Table 6 for the other vehicle sizes and road types.  Table 6 for the other vehicle sizes and road types.  Table 6 for the other vehicle sizes and road types.  Table 6 for the other vehicle sizes and road types.  Table 6 for the other vehicle sizes and road types.  Table 6 for the other vehicle sizes and road types.  Table 6 for the other vehicle sizes and road types.  Table 6 for the other vehicle sizes and road types.  Table 6 for the other vehicle sizes and road types.  Table 6 for the other vehicle sizes and road types.  Table 6 for the other vehicle sizes and road types.  Table 6 for the other vehicle sizes and road types.  Table 6 for the other vehicle sizes and road types.  Table 6 for the other vehicle sizes and road types.  Table 6 for the other vehicle sizes and road types.  Table 6 for the other vehicle sizes and road types.  Table 6 for the other vehicle sizes and road types.  Table 6 for the other vehicle sizes and road types.  Table 6 for the other vehicle sizes and road types.  Table 6 for the other vehicle sizes and road types.  Table 6 for the other vehicle sizes and road types.  Table 6 for the other vehicle sizes and road types.  Table 6 for the other vehicle sizes and road types.  Table 6 for the other vehicle sizes and road types.  Table 6 for the other vehicle sizes and road types.  Table 6 for the other vehicle sizes and road types.  Table 6 for the other vehicle sizes and road types.  Table 6 for the other vehicle sizes and road types.  Table 6 for the other vehicle sizes and road types.  Table 6 for the other vehicle sizes and road types.  Table 6 for the other vehicle sizes and road types.  Table 6 for the other vehicle sizes and road types.  Table 6 for the other vehicle sizes and road types.  Table 6 for the other vehicle sizes and road types.  Table 6 for the other vehicle sizes and road types.  Table 6 for the other vehicle sizes and road types.  Table 6 for the other vehicle sizes and road types.  Table 6 for the other vehicle sizes and road types.  Table 6 for the other vehicle sizes and road types.  Table 6 for the other vehicle sizes and road types.  Table 6 for the other vehicle sizes and road types.  Table 6 for the other vehicle sizes and road types.  Table 6 for the other vehicle sizes and road types.  Table 6 for the other vehicle sizes and road types.  Table 6 for the other vehicle sizes and road types.  Table 6 for the other vehicle sizes and road types.  Table 6 for the other vehicle sizes and road types.  Table 6 for the other vehicle sizes and road types.  Table 6 for the other vehicle sizes and road types.  Table 6 for the other vehicle sizes and road types.  Table 6 for the other vehicle sizes and road types.  Table 6 for the other vehicle sizes and road types.  Table 6 for the other vehicle sizes and road types.  Table 6 for the other vehicle sizes and road types.  Table 6 for the other vehicle sizes and road types.  Table 6 for the other vehicle sizes and road types.  Table 6 for the other vehicle sizes and road types.  Table 6 for the other vehicle sizes and road types.  Table 6 for the other vehicle sizes and road types.  Table 6 for the other vehicle sizes and road types.  Table 6 for the other vehicle sizes and road types.  Table 6 for the other vehicle sizes and road types.  Table 6 for the other vehicle sizes and road types.  Table 6 for the other vehicle sizes and road types.  Table 6 for the other vehicle sizes and road types.  Table 6 for the other vehicle sizes and road types.  Table 6 for the other vehicle sizes and road types.  Table 6 for the other vehicle sizes and road types.  Table 6 for the other vehicle sizes and road types.  Table 6 for the other vehicle sizes and road types.  Table 6 for the other vehicle sizes and road types.  Table 6 for the other vehicle sizes and road types.  Table 6 for the other vehicle sizes and road types.  Table 6 for the other vehicle sizes and road types.  Table 6 for the other vehicle sizes and road types.  Table 6 for the other vehicle sizes and road types.  Table 6 for the other vehicle sizes and road types.  Table 6 for the other vehicle sizes and road types.  Table 6 for the other vehicle sizes and road types.  Table 6 for the other vehicle sizes and road types.  Table 6 for the other vehicle sizes and road types.  Table 6 for the other vehicle sizes and road types.  Table 6 for the other vehicle sizes and road types.  Table 6 for the other vehicle sizes and road types.  Table 6 for the other vehicle sizes and road types.  Table 6 for the other vehicle sizes and road types.  Table 6 for the other vehicle sizes and road types.  Table 6 for the other vehicle sizes and road types.  Table 6 for the other vehicle sizes and road types.  Table 6 for the other vehicle sizes and road types.  Table 6 for the other vehicle sizes and road types.  Table 6 for the other vehicle sizes and road types.  Table 6 for the other vehicle sizes and road types.  Table 6 for the other vehicle sizes and road types.  Table 6 for the other vehicle sizes and road types.  Table 6 for the other vehicle sizes and road types.  Table 6 for the other vehicle sizes and road types.  Table 6 for the other vehicle sizes and road types.  Table 6 for the other vehicle sizes and road types.  Table 6 for the other vehicle sizes and road types.  Table 6 for the other vehicle sizes and road types.