Can product service systems support electric vehicle adoption?

Plug-in electric vehicles are seen as a promising option to reduce oil dependency, greenhouse gas emissions, particulate matter pollution, nitrogen oxide emissions and noise caused by individual road transportation. But how is it possible to foster diffusion of plug-in electric vehicles? Our research focuses on the question whether e-mobility product service systems (i.e. plug-in electric vehicles, interconnected charging infrastructure as well as charging platform and additional services) are supportive to plug-in electric vehicle adoption in professional environments. Our user oriented techno-economic analysis of costs and benefits is based on empirical data originating from 109 organizational fleets participating in a field trial in south-west Germany with in total 327 plug-in electric vehicles and 181 charging points. The results show that organizations indicate a high willingness to pay for e-mobility product service systems. Organizations encounter non-monetary benefits, which on average overcompensate the current higher total cost of ownership of plug-in electric vehicles compared to internal combustion engine vehicles. However, the willingness to pay for e-mobility charging infrastructure and services alone is currently not sufficient to cover corresponding actual costs. The paper relates the interconnected charging infrastructure solutions under study to the development of the internet of things and smarter cities and draws implications on this development.


Introduction
Increasing awareness of the transport sector's significant contribution to climate change, oil dependency, particulate matter pollution, nitrogen oxide emissions and noise particularly in urban areas has resulted in activities for road transport electrification. Substituting internal combustion engine vehicles (ICEV) with plug-in electric vehicles (EV), i.e. full battery electric vehicles (BEV), range extended electric vehicles (REEV) and plug-in hybrid electric vehicles (PHEV), seems a very promising step to cope with the challenges of individual road transport and fits to the smart city paradigm, which has become one of the most important urban strategies to foster green growth and to improve urban sustainability against the backdrop of climate change (March, 2016). A wide range of definitions for the fuzzy smart city paradigm exist (Cocchia, 2014). Despite the risks accompanying hyper-connected societies (Rifkin, 2014), the common notion is that in smart cities information and communication technologies (ICT) are used to increase citizens' quality of life while contributing to sustainability (Cocchia, 2014;Yeh, 2017). Highly connected information systems providing real-time digital platform services connecting citizens with urban infrastructures are key resources for smart cities.
Policy incentives and car manufacturers' portfolio decisions have a positive influence on EV adoption (Langbroek et al., 2016). Consequently, registrations of EV have been continuously increasing in industrialized countries on the global scale since 2008 (IEA, https://doi.org/10.1016/j.tra.2018.04.028

Methods and data
In order to evaluate e-mobility PSS in organizational fleets (i.e. EV, EVSE and e-mobility services) cost-benefit analysis is applied. Organization-specific costs per vehicle for using e-mobility PSS are assessed and compared to the organization-specific WTPM in order to calculate corresponding net benefits. Section 2.1 describes the methodological framework to evaluate costs and benefits of the PSS. Section 2.2 describes the data collected during an e-mobility project in south-west Germany between 2013 and 2015. The project included a large-scale fleet trial with 109 organizations owning 327 EV as well as a regional charging network with 181 interconnected charging points (Sachs et al., 2016).

Framework to evaluate costs and benefits of PSS
We now turn to the methodology, which is used to evaluate net benefits of e-mobility PSS. As costs for EV and ICEV differ, we analyze the difference of costs and a potential WTPM. Data on costs are combined with data on user perceptions by applying costbenefit analysis based on the following main Eq. (1): Net benefits of PSS u i PSS are determined for every organization i by adding net benefits of EV u i EV , EVSE u i EVSE , hardware connectivity services u i CON , service bundles u i SB (cf. While all cost parameters are determined by average cost values derived from scientific studies (Plötz et al., 2013), parameters on WTPM are survey based. All parameters have equal weights as WTPM, costs for the vehicles, EVSE and services have monetary units that are directly capable of being totaled.
The average TCO difference between EV and ICEV for a company is calculated in Eq. (3): Here, the difference between the cheapest EV (BEV, PHEV) and the cheapest ICEV (Gasoline, Diesel) is determined for all vehicles considered of being substituted by EV k of company i, then summed up and divided by their number K (Eq. (3)). For all vehicles K , TCO is estimated for all vehicle types by Eq. (4).

=+
Here, I r veh is the vehicle purchase price and SP r veh equals the residual value for the vehicle at the end of usage time T veh . The formula is completed by the interest rateĩ . Residual values for the vehicle at end of use T veh , SP r veh are determined as in Plötz et al. (2014). Operating expenditures are calculated as depicted in Eq. (6).
A part of the operating expenditures is mileage dependent. Annual vehicle kilometers travelled by vehicle k (VKT k ) are derived from vehicle driving profiles collected with GPS-trackers during this study's field trial (all trips of a vehicle within a certain observation period, cf. Section 2.2). The range of an EV determines the feasibility of trips that can be covered by BEV and the electric driving share of PHEV. This largely affects the economics compared to ICEV. It is thus necessary to consider individual driving (e.g. Gnann et al., 2015a). In order to calculate operating expenditures (a r opex ) annual vehicle kilometers travelled (VKT ) k , are multiplied with mileage dependent costs such as fuel costs for electricity or conventional fuels (Eq. (6) (6)). While the annual vehicle kilometers travelled are estimated based on an extrapolation of the distance driven in the driving profile, the share of electric driving of a PHEV is determined in an EV simulation (Plötz et al., 2014). All cost parameters considered are taken from Plötz et al. (2014) and Gnann et al. (2015a). We consider an interest rateĩ of 5%  and an average vehicle holding time for new vehicles of 3.8 years (KBA; Gnann et al., 2015a) which will be considered equal to the investment horizon T veh . The assumptions for battery and energy prices are shown in Table A (7)): In addition to the hardware connectivity services s CON different e-mobility services compose two different service bundles, SB1 and SB2 (Table 1). Customized service bundles SBopt i are constructed in addition, i.e. service bundles consisting only of charging platform and additional services with Hardware connectivity services s CON grant permission to the charging network and guarantee EVSE connectivity and maintenance. SB1 represents the service bundle which was in addition to the hardware connectivity services s CON provided in the field trial (Table 1). It is composed of the following charging platform services: a map-based display of charging points and display concerning their current availability (s 1 ), a billing service for charging at the own organizations' EVSE (s 3 ), a service managing the provision of own EVSE in a regional charging network (s a 4 ) and a service managing charging activities at other organizations' EVSE within a regional charging network (s a 5 ). Furthermore, consultancy services for EV and EVSE (s a 7 ) complete SB1. More detailed information is available in Sachs et al. (2016). SB2 represents a fictitious, more extensive service bundle that could potentially be provided to emobility PSS adopters in the future (Table 1). In addition to the services of SB1, this service bundle includes the possibility to reserve charging points (s 2 ), provides access to a supra-regional charging network (s b 4 and s b 5 ), includes a mobility guarantee (s 6 ) and consultancy services by one provider instead of several different providers (s b 7 ). A detailed description on how the relevant services of this study were chosen by Ensslen et al. (2017).
WTPM for the service bundles wtpm i SB are calculated based on WTPM for single service components wtpm i s using linear, additive utility functions (Weddeling et al., 2010). WTPM for single services are aggregated in order to calculate WTPM and net benefits of the service bundles considered.
Operating expenditures are calculated by summing up service costs, i.e. adding cost of service bundle c i SB and costs for connectivity c i CON . Parameters and equations for the calculations of costs for EVSE and services are provided in Table A.3. With the additional information on organizations' benefits available, i.e. the survey-based information on WTPM for EV, EVSE and e-mobility services as well as the data on the compensation of expenses granted wtpm wtpm wtpm wtpm COE ( ,,,

Description of datasets
All organizations interested to get involved in the project's field trial were asked to participate in a voluntary fleet analysis involving a detailed analysis of driving patterns by logging driving profiles in order to find out about their fleets' electrification potentials. 45 of the interested 234 organizations volunteered to participate. 109 participated in the field trial, i.e. decided to purchase at least one EV, project specific EVSE as well as additional hardware connectivity services. 26 organizations are represented in both subsamples, i.e. provided driving profile data and participated in the field trial ( Fig. 1).
As this study focuses on early EV adopting organizations we did not expect the organizations' distributions concerning industrial Map-based display of charging points and display concerning their current availability s 1 xx Reservation of charging points s 2 x Billing service for charging at the own EVSE s 3 xx Provision of own EVSE…… in a regional charging network s a 4 xx … in a supra-regional charging network s b 4 x Usage of other organizations' EVSE…… within a regional charging network s a 5 xx … within a supra-regional charging network x Additional services Mobility guarantee (a rental car is delivered to the workplace or to the employees if needed) s 6 x EV and infrastructure consulting by …… different providers s a 7 x … the same provider s b 7 x A. Ensslen et al. Transportation Research Part A xxx (xxxx) xxx-xxx sectors and size to be representative for south-west Germany. We are positively surprised that industrial sector distributions of participating organizations represent the number of employees in the different industrial sectors in Baden-Württemberg fairly well. 38% of the participating organizations belong to the manufacturing sector, 14% to the wholesale and retail sector, 13% to public administration, 8% to information and communication and 6% to the construction sector. Less than 5% of the population are represented in the remaining sectors (Table B.1). Differences can be observed concerning organizations' sectoral distributions compared to Germany's new car registrations (Table B.1). About 80% of the organizations in this study employ less than 250 persons, i.e. are small and medium-sized enterprises. This share is comparably high, as only about 50% of employees in Baden-Württemberg work for organizations employing less than 250 persons (Table B.2).
The fleet managers and decision makers in the participating organizations are on average 45 years old (SD = 12), are predominantly male (about 85%) and are well educated. About half of them have completed academic studies and about 30% have a degree at university entrance level or a master craftsman diploma. 50% have a technical, about 40% a commercial background. On average, the respondents have been employed for 16 years in their organizations (SD = 12) and have an experience level with fleet management activities of 10 years on average (SD = 10). Half of them dedicate more than 10 h per month to fleet management activities, 25% four hours or less and 25% more than 20 h .
Field trial participation came along with a monthly compensation of expenses. Constrained by the real costs for EV the participating organizations received up to 500 Euros monthly (w/o VAT) per BEV or REEV and 350 Euros monthly (w/o VAT) per PHEV in order to compensate for additional costs of project specific EVSE, for the still existent economic disadvantages of EV and for providing data. Due to this setup collecting the substantial amount of high quality organization-specific information was possible.
An overview on the specific datasets used for the assessment of WTPM, costs and net benefits of EV, EVSE and the services considered is provided in Table C.1.
The vehicle driving profile dataset is available for the subsample of 45 organizations, who have not necessarily decided to participate in the field trial at the point of time the data was collected. The dataset consists of profiles of all trips of a vehicle within at least three weeks of observation which were collected with GPS-trackers for ICEV to test potential replacements by EV. Several information about the company (such as the number of employees, the main company site or the size of the city) as well as the vehicle (vehicle size, usage type, main overnight parking or number of users) was collected in a short survey. This dataset contains driving profiles of 223 commercially licensed vehicles of 45 organizations participating in the project. The 223 vehicle driving profiles were collected over an average observation period of 23.1 days (SD: 5.8 days) with an average daily mileage of 56.6 km (SD: 39.7 km). At night, the vehicles are mainly parked at the company site (90%) with a dedicated parking spot (54%). Most of the vehicles are fleet vehicles which are used by several users (70%) ( Table B.3). Similar to the sample of organizations participating in the field trial, most of the 45 companies which volunteered to provide GPS tracks have less than 250 employees and are mostly located in small cities below 100,000 inhabitants.
Survey data on WTPM for EV, EVSE and e-mobility services was collected between five and nine months before the end of the field trial (between April and August 2015). Organizations had experienced e-mobility PSS for quite some time. Survey data includes organization specific information on WTPM for EV wtpm i EV , EVSE wtpm i EVSE , basic hardware connectivity services wtpm i CON as well as charging platform and additional services s, wtpm i s . The following survey questions were answered by the organizations participating in the field trial: (i) wtpm i EV : "How much would your organization be willing to pay for a BEV and a PHEV?" (ii) wtpm i EVSE : "Which monthly extra charge (price reduction) would your organization be willing to pay (expect) for the nonmonetary characteristics of a charging station?" (iii) wtpm i s :" What is the maximum price that your organization is willing to pay for the following services?" The detailed questions asked in the survey are provided in Appendix E.
Invoices of project specific interconnected EVSE installed were collected between May 2014 and July 2015 by mail. A. Ensslen et al. Transportation Research Part A xxx (xxxx) xxx-xxx Information available on organizations' investments for project specific interconnected EVSE (I i EVSE ) is then used to calculate costs of project specific EVSE TCO i EVSE .
Different cost parameters are used for EV and service specific cost calculations (Appendix A). As the e-mobility services considered were partly tested within the project, actual cost parameters for a large part of the services (s CON , s 1 , s 3 , s a 4 , s a 5 ) are used. For the services not actively tested within the project (sss ,, bb 456 ) or free of charge within the project context (ss , 2 7 ) plausible assumption were made during a stakeholder workshop .

Results
In order to evaluate costs and benefits of e-mobility PSS we first analyze EV (Section 3.1). Second, we analyze EVSE and corresponding connected e-mobility services (Section 3.2). Third, the results concerning costs and benefits of EV, EVSE and e-mobility services are combined in order to calculate net benefits of the whole e-mobility PSS (Section 3.3). Sensitivity analysis are conducted to analyze effects of parameter variations on TCO and overall net benefits (Section 3.4).

Costs, WTPM and net benefits of EV
We first have a look at the TCO differences of the vehicles (Eq. (2)). For this reason, we compare annual TCO for the cheapest EV with those of the cheapest ICEV. We observe that in 2015, annual TCO are about 800 € higher for most users (∼70%) and hardly amortizable with current prices and driving behavior (Table E.1). Thus in 2015, based on the driving profiles analyzed EV cannot be paid off. Positive TCO seem only possible in selected use cases (Schücking et al., 2017). These cost differences do not include the costs for EVSE which might be considerable (Section 3.2). However, the significant additional EV specific costs might be compensated by companies' WTPM for an environmental or marketing effect.
We therefore analyze stated price premiums (WTPM) organizations are willing to pay for non-monetary values of EV based on survey data described in Section 2.2. wtpm i EV which is representing annual WTPM for the organizations' EV is calculated by multiplying the organizations' EV specific WTPM with the annuity factor. According to the results, a large part of the organizations participating in the field trial is willing to pay considerably more for an EV than for an ICEV. About 50% of the organizations are willing to pay more than 5000 € additionally for an EV compared to the respective ICEV, i.e. 1477 annually. 75% of them are willing to pay at least a premium of 2000 €, i.e. 591 € annually (Table E.1). These results point out that (early) commercial EV adopters are likely to be willing to pay considerably more for EV beyond what is expected from a pure economic point of view.
Comparing costs and stated WTPM for EV shows that on average WTPM for EV is higher than the additional costs of EV compared to ICEV ( > wtpm TCO Δ i EV i EV ) resulting in positive net benefits (Table E.1). The advantages of EV seem to outweigh corresponding disadvantages, particularly unfavorable TCO differences, range limitations and high charging times.

Costs, WTPM and net benefits of EVSE and e-mobility services
Results on organizations' costs for EVSE and e-mobility services are presented in Table E are 323 €/a. As net benefits of different charging platform and additional services are negative for many organizations, they are not considered in SBopt i . TCO i EVSE represent the major cost component of products and services additional to EV. The costs for connectivity services (c i CON ), charging platform and additional services provided within the project (SB1) are comparatively low compared to the annual amortization rates for EVSE. This is also the case for SB2. However, according to the results concerning customized service bundles SBopt i , costs for services should be considerably reduced, as only some of the services show positive net benefits for the users.
The results on surveyed WTPM for EVSE and e-mobility services (Table E.2) show that the field trial participants are on average willing to pay an annual surcharge of 789 €/a for the non-monetary values of EVSE (wtpm i EVSE ). On average organizations are willing to pay 205 €/a for connectivity services (wtpm i CON ). Results concerning WTPM for charging platform and additional services composing the service bundles considered show that the organizations are willing to pay somewhat more for SB2 (on average 913 €/a) than for SB1 (on average 725 €/a) and SBopt i (on average 763 €/a).
Net benefits of EVSE and e-mobility services are negative for most organizations (Table E.2). Average net benefits of connectivity services (u i CON ) and customized service bundles (SBopt i ) are positive. About 85% of the organizations show positive net benefits for the customized service bundles SBopt i . However, net benefits of EVSE and corresponding services ( ++ u uu i EVSE i CON i SB ) without considering EV are negative for between 75% and 85% of the organizations (Table E.2). Net benefits of EVSE and organization specific customized service bundles show higher net benefits than the pre-defined service bundles (Fig. 2).

Net benefits of e-mobility PSS
After analyzing components of the PSS individually, net benefits of e-mobility PSS are now evaluated as a whole. The monetary incentives granted are also considered. According to Table E The standard errors (SE) of the expected values of the whole PSS are calculated as follows (Eq. (9) ). Expected net benefits of the e-mobility PSS without considering monetary incentives are 126 €/a. Adding the expected value of the compensation of expenses per EV (2050 €/a) results in expected positive net benefits of 2176 €/a. Hence, negative net benefits of the e-mobility PSS of many organization are largely overcompensated by the monetary incentives provided (Fig. 3).

Sensitivity analysis
Our sensitivity analysis focuses on parameters potentially influencing TCO calculations as corresponding cost parameters might heavily change in the future and regional aspects might influence some of the parameters. Fig. 4 shows the effects of parameter variation on TCO calculations and consequently e-mobility PSS specific net benefits. Sensitivities within the range of 25% are highest for battery price, followed by interest rate, fuel prices and electricity price.

Discussion
This study deals with the analysis and evaluation of e-mobility PSS based on an approach that combines techno-economic and user-behavioral aspects in order to answer the central research question whether e-mobility PSS in company fleets can support EV adoption. Section 4.1 discusses and critically reflects methodological aspects of our research. In Section 4.2 results are discussed.

Discussion of methodology
In order to combine the techno-economic analysis of e-mobility PSS with non-monetary benefits, we conducted a survey with participants of a field trial with EV and let them estimate the non-monetary value of e-mobility PSS including different service bundles. The TCO approach is often criticized as being inconclusive for a vehicle purchase decision (ESMT, 2011;McKinsey, 2010;Mock, 2010;NPE, 2011;Peters et al., 2012;Plötz et al., 2012;Thiel et al., 2010), yet it was repeatedly stated by organizations that it is the most important aspect in a commercial vehicle buying decision (Dataforce, 2011). However, this does not necessarily mean that organizations really calculate TCO in their vehicle buying decisions. They might rather use perceived estimates in their decision-making processes. Though the vehicle buying decision is complex  and may include several decision making steps (Klöckner, 2014), the focus of this study is to analyze net benefits of e-mobility PSS as a whole. Therefore, the approach applied combining techno-economic assessments with behavioral aspects seems reasonable.
The design of the questionnaire allowed the fleet managers and decision makers of commercial vehicle users to set non-monetary values of EV, EVSE and e-mobility services in relation to their actual costs. As we asked the fleet managers about WTPM for EV, EVSE, and corresponding services, they tried to monetarize the benefits. However, do these monetarized benefits appropriately represent the real benefits of the e-mobility PSS? The large spread in the net benefits (Fig. 3) shows that corresponding perceptions vary significantly between organizations. Consequently, the results should be interpreted carefully.
The participating organizations represent a very special early adopter group that received expense compensations for participating. The survey sample consists of early EV adopters so it is difficult to draw conclusions about future adopters of EV who will enter the market later and might have different motivations. For example, the average WTPM for e-mobility PSS might be lower when an early majority (Rogers, 1962) is about to enter the market and it is also expected to decrease with increasing market diffusion (Gölz et al., 2015). Hence, results should be considered as an upper estimate. However, some services could potentially increase their attractiveness with increasing market penetrations (e.g. due to a higher number of charging stations) potentially resulting in increasing WTPM.
Furthermore, in addition to incomplete datasets due to the different subsamples of different datasets, missing value problems reduced sample sizes (Appendix C). We controlled for potential errors by additionally calculating net benefit based on the subsample with full data availability. Differences observed between the two approaches are not significant.
Costs for charging infrastructure and services were taken from an early stage e-mobility project. The participating organizations' decisions to adopt and use the interconnected charging infrastructure and services might also be linked to an increased WTPM. A. Ensslen et al. Transportation Research Part A xxx (xxxx) xxx-xxx However, the comparably high costs for EVSE and e-mobility services provided within this project might represent the current market situation.
Our TCO and e-mobility PSS net benefit calculations are based on a large set of parameters and assumptions. These are related to values observed in 2015, the observation country (Germany) and the specific observation population (early EV adopting organizations). In the future preconditions for e-mobility PSS specific net benefits might change increasing net benefits of EV. Battery prices are assumed to continue to decline (Berckmans et al., 2017;IEA, 2017). Second hand values of diesel and gasoline cars might decrease quicker than assumed due to governments' plans to ban ICEV from many cities (IEA, 2017). Although the short investment horizon of 3.8 years used in this study represents the average holding time of cars in the commercial sector of Germany, strong variations depending on organizations' commercial sector are possible (Gnann et al., 2015a). According to analyses of other studies, long investment horizons are in favor of EV and sensitivities are comparably low (Palmer et al., 2018). Furthermore, electricity and gasoline prices as well as incentive schemes differ between countries and regions and should therefore be considered when interpreting the results presented.

Discussion of results
The results presented in Section 3.1 indicate that despite the disadvantages of EV (e.g. range limitations, high charging times) WTPM for EV of organizations participating in the field trial compensate additional EV specific costs. These findings are in line with Plötz et al. (2014) and Peters and Dütschke (2014) showing that private EV users are willing to pay more for EV compared to ICEV. WTPM for EV can be explained by innovative car pool managers benefitting from EV procurements due to technophilia . Furthermore, organizations benefit from positive effects on employee motivations  and EVs' innovative and environmental image (Guth et al., 2017;Nesbitt and Davies, 2013;Sierzchula, 2014).
The results presented in Section 3.2 indicate that net benefits of individually customized service bundles are higher than net benefits of the predefined ones (Fig. 2). Individual consulting for individual composition of service bundles could increase corresponding net benefits significantly. The market for e-mobility PSS is still highly diversified. Hence, requirements for individual consultancy services are high. There might not be only one e-mobility PSS to succeed in sales activities in organizational fleets.
Creating customized e-mobility PSS offers for different types of potential EV adopting organizations might be a convincing strategy for stakeholders and would fit with the smart city paradigm assuming a higher flexibility of services to accommodate individual needs. As interconnected EVSE of other organizations were used only infrequently in our case study, inter-organizational charging activities could hardly be observed. Nevertheless, it seems that the users interpret this service as a kind of insurance against flat batteries and are therefore willing to pay for this charging platform service even without its actual usage. Furthermore, platformconnected EVSE delivers a sound basis for providing smart energy services Ensslen et al., 2018;Goebel et al., 2014;Salah et al., 2017) and multimodal platform services (Willing et al., 2017) including offerings as e.g. ride sharing (Teubner and Flath, 2015) and corporate carsharing (Heinen and Pöppelbu, 2017) in addition to the connected charging platform services considered in this case study. Such additional platform services are co-creating value for EV users and providers by using information exchanged in real-time between EV, EV users, service platforms and other stakeholders. This could contribute to balance the negative benefits of EVSE.
The results presented in Section 3.3 show that today financial support can be an important incentive for EV adoption. If prices for EV, EVSE and e-mobility services are further decreasing, monetary incentives could also be reduced. The findings of this paper show that annual net benefits for most organizations are clearly positive due to the compensation of expenses granted. For about half of the organizations net benefits of the EV are positive without considering the effect of the monetary incentives. However, net benefits of interconnected EVSE and corresponding services are negative for about 80% of the participating organizations. Sierzchula et al. (2014) as well as Harryson et al. (2015) show that financial incentives and availability of EVSE are positively correlated with different countries' EV market shares. The results of this case study point out that the diffusion of EV could be supported not only by providing incentives to vehicle acquisitions but also by incentivizing e-mobility PSS including interconnected EVSE and corresponding platform services being part of publicly accessible charging networks. This could result in positive spillover effects, as additional publicly accessible EVSE offering smart charging services would be put in place that would again positively impact EV sales and developments towards smarter mobility solutions and be in line with the development towards the smart city paradigm.
The results of our sensitivity analysis (Section 3.4) show that EVs' TCO (and therefore overall net benefits of e-mobility PSS) are particularly sensitive to variations of battery prices. Expected fast decreasing battery prices more than halving until 2020 (Berckmans et al., 2017) would significantly increase net benefits of EV (∼475 €/a). Effects of electricity and fuel price parameter changes within the range of 25% are comparably low. Despite the high sensitivity potential of future battery price developments, the lever of governments' incentive programs on overall net benefits of e-mobility PSS is comparably high. Incentives amount to more than 2500 €/a in Norway, more than 2000 €/a in France (Harryson et al., 2015) and to more than 2000 €/a in this project's fleet test. These findings are in line with Palmer et al. (2018) showing that government support for low-emission vehicles clearly needs to address financial barriers if EV market share is to break out of the niche market.

Summary, conclusions and future work
Recently many field trials with EV intending to counteract climate change, to reduce oil dependency, particulate matter pollution and noise emissions in urban areas by electrifying road transport were carried out in order to develop corresponding technologies. During a field trial with 109 organizations using 327 EV driving profiles, survey data, actual costs for interconnected EVSE solutions A. Ensslen et al. Transportation Research Part A xxx (xxxx) xxx-xxx as well as information on the compensation of expenses granted to organizations participating were collected. Net benefits for e-mobility PSS, i.e. EV, interconnected EVSE and e-mobility services were evaluated based on fleet managers' perspectives by analyzing costs and WTPM. The central research question addressed in this article whether e-mobility PSS can support EV adoption can be answered as follows: Currently the costs for interconnected EVSE solutions and e-mobility services outweigh corresponding WTPM. However, e-mobility PSS offerings are more promising if they are adapted to individual needs. They might become even more beneficial to EV users, particularly by considering benefits of smart energy services and multimodal platform services in addition. Consequently, WTPM for e-mobility PSS could increase resulting in a higher probability to adopt. In addition to that it is very likely that interconnected EVSE solutions and corresponding services allocated on the market underlie economies of scale and so will become cheaper in the future. Consequently, positive net benefits of e-mobility PSS might be possible for more organizations in the future without government incentives, particularly if increasing inter-organizational usage frequencies of EVSE are taken into account. Therefore, it is very likely that e-mobility PSS will directly support EV adoption in the future. However, in the current market phase the EVSE and e-mobility services offered rather negatively affected the adoption of EV, the high prices of interconnected EVSE in particular. Although the EVSE and e-mobility services offered did not directly contribute positively to higher net benefits of emobility PSS in most organizations, extending the e-mobility charging service offering by further additional smart platform services following the smart city paradigm might positively affect overall net benefits. Smart energy demand response services, billing services permitting to charge private EV with photovoltaic energy produced at the home roof top at the workplace, billing services to charge company cars at home with electricity paid by the employer and multimodal platform services are additional services that could enhance e-mobility PSS offerings. In addition, positive effects of publicly accessible EVSE encountering range anxiety should be considered before conclusions are made concerning the research question whether e-mobility PSS can support EV adoption. Considering positive indirect effects of EVSE availability on EV diffusion should be particularly considered when incentive schemes are designed. The financial incentive program of this study's field trial supported PSS sales activities, i.e. to allocate EV, project specific interconnected EVSE and corresponding charging platform services. This resulted in positive overall net benefits for most of the participating organizations.
Future work could in addition to services considered in this analysis focus on further additional, customer-oriented smart services, as interconnected EVSE solutions provide the basis for EV specific EVSE being part of the internet of things fostering possibilities to offer further e-mobility specific charging platform services to organizations and EV users. Future work could focus on evaluating costs and benefits of such advanced e-mobility PSS integrating additional innovative charging platform services forming service bundles supportive to the smart city paradigm.