Measures to resolve range anxiety in electric vehicle users

Abstract

Electric Vehicle (EV) technology is the key to emission reduction and energy efficiency improvement targets while decreasing the dependence on fossil fuel. Despite the fact that replacing conventional vehicle powered by internal combustion engine with EVs reduce greenhouse gas emissions, limited range of EVs has been leading their mass deployment towards saturation. This paper performs a scoping review of articles on range anxiety induced among EV users. It explores user’s perception and actual affecting factors on range of EVs, and proceeds on to measures suggested in the articles to mitigate range anxiety.

1 INTRODUCTION

Electric Vehicle (EV) technology is of topical concern in the contemporary world owing to the fact that it is the key to emission reduction and energy efficiency improvement targets in relation to transportation sector. Heightening concerns of global warming and exponential decrease of energy returned on energy invested of petroleum products have obligated the world to opt for sustainable greener technologies, such as EVs, to possibly revert climate change [1, 2]. The environmental benefits further accelerates if EVs are powered using cleaner sources of energy [3]. But, unaccustomed state-of-the-art technologies, such as EVs, are treated as an inferior to dominant technologies such as Internal Combustion Engine Vehicles (ICEVs) [4]. EVs, referring to Battery Electric Vehicles in this study, are powered exclusively by batteries. Lithium ion batteries used in EVs have almost 25 times lower energy density compared to gasoline used in ICEVs [5]. Due to this reason, the range, which is the distance a vehicle can provide when fully fueled or charged if driven under normal conditions, is limited for EVs compared to ICEVs [6, 7]. The relatively limited range gives rise to a psychological phenomenon called range anxiety in EV users, which is labelled as one of the most crucial barriers in EV adoption, specifically during early adoption phase, eventually hindering the utility of EVs. Overcoming this fear is mandatory for substantial diffusion of EVs in the emerging market [8].

While improvement in battery technology will increase the range of EVs and serve to diminish range anxiety in the future, researchers have studied the specifics of range anxiety and come across variety of solutions proposed and applied in diverse regions. The approach differs depending upon the economic status of a country and its level of market [9]. This paper performs scoping review of literatures on range anxiety in EV users to find possible measures to resolve it, discusses the relevance of its different aspects to each of the identified measures and suggests possible chronology of undertaking the measures for an emerging EV market scenario.

2 METHOD

Research articles related to range anxiety are identified by establishing the following research questions:

• Is range anxiety logical or far-fetched?

• What are the factors that actually affect the range of an EV?

• What are the solutions to mitigate range anxiety?

The keywords used to sort out the articles are ‘Electric Vehicles’ and ‘Range Anxiety’. The article selection flow chart for scoping review performed in this paper is given in Figure 1.

The following sections of the paper are user’s perception on range anxiety and actual factors affecting the range, mitigating measures to resolve range anxiety, relevance of identified measures with different aspects of range anxiety, suggested chronology of application of identified measures for an emerging EV market and conclusion.

Figure 1

Article Selection Flow Chart

3 USER’S PERCEPTION ON RANGE ANXIETY AND ACTUAL FACTORS AFFECTING EV RANGE

User’s cognitive perception on a new technology is driven by juxtaposition with the conventional ones, and plays a considerable role in making preference [10]. ICEVs, with sufficing energy density to meet user’s need, have been ruling automotive world for centuries. EVs, on the other hand, provide relatively limited range, and there has not been enough systematic scrutinization on whether or to what extent they satisfy the user’s demand.

While some research claim that today’s EVs are fitting enough to be used for daily needs, some of them claim otherwise. As per a research done in 2013, EVs can offer 100 miles on a single charge in affordable price [8] and for everyday range close to this value, another study based on US drivers shows that it is suitable to switch to EVs [11]. Present day EVs offer range over 100 km as shown by a study done in Sweden in 2022 which indicates a 133-km range BEV as short-range EVs [12]. Similar is the case with Finland and Switzerland where EVs can cover 85–90% of all national trips as of 2016 [13]. Another research shows that a 75-mile range EV can satisfy the requirement of 86–90% of the desired miles with level 1 or home-based charging facility [3]. Nevertheless, fear of being stranded prompts the users to opt for higher range EVs or ICEVs. In an experiment done to grasp user’s attitude towards EVs, in which a 120 km range electric car was used, the participants expected to use a higher range ones just so as to feel more comfortable [14]. The uncertainty is, however, regarded reasonable by an investigation done in main high-tech and financial areas of Beijing which shows that during weekdays only 46% drivers’ requirement are fulfilled by EVs [15]. Such failure likely calls for a change in daily activities of drivers for which everyone might not be comfortable with.

Aside from variation in offered range, information on the route to be travelled, reliability on the available range estimated for the vehicle, and confidence and skill to operate the vehicle also escalate or diminish range anxiety and ultimately determine propensity towards EVs [16]. Merely comparing EVs and ICEVs is not adequate in gaining complete user’s perspective [10]. A study revealed that psychological concerns associated with range anxiety arises at about 15 km of remaining driving range [17]. Experienced EV users having range related knowledge are found to be less sensitive about such circumstances compared to the unexperienced ones [18, 19]. It has been confirmed by an experiment that recreated a critical range situation in which the drivers had experienced comparatively less range anxiety after having travelled through a trip [20]. It implies that when the users are made familiar with their actual travel pattern, the level of acceptance increases. But inadequate exposure to travel requirements and vehicle capabilities leads to high level of rejection [10]. Strategies to enhance range experience include imparting relevant knowledge before purchase, and during usage [19, 21]. Aside from this, understanding how transport and energy systems influence EV driving practices is also important [22].

While lack of confidence for adopting EVs in relation to limited range might be rational, a study accuses such sentiment being an excuse to refrain from changing their behavior. This explanation was given on the basis of study on failure of an EV company that operated in Denmark and Israel [23]. Further bolstering this assertion, another study has declared range anxiety as a rhetorical excuse [24].

In real-world situation, range of an EV depends upon driving environment aside from inherent technology. A ‘stop and go’ driving cycle consumes relatively higher energy compared to highway driving cycle due to constant demand for acceleration. Also, the faster the acceleration the higher the energy consumed. Similarly, driving on an uphill gradient requires more energy in relation to driving on flat roads or downhill [25]. Ambient temperature is another influencing factor because it determines the use of auxiliary devices and output energy losses, thus affecting the energy consumption [26]. Higher temperature is accompanied by increase in total energy consumption [27]. Optimal driving speed of an EV at certain temperature increases with an increase in the ambient temperature [28]. EV range also depends on rolling resistance, aerodynamic drag, battery chemistry and energy density [29].

4 MITIGATING MEASURES TO RESOLVE RANGE ANXIETY

Studies have pointed out some measures to address range anxiety. Majority of the researchers have emphasized on building different levels and types of charging infrastructures on various locations. Aside from that, accurate estimation of range and ways of optimizing energy consumption for added range are also studies. These three major mitigating measures along with the others are described below.

4.1 Building charging infrastructures

The cost penalty perceived at the time of purchase given the limited range and long recharging time of an EV is substantially addressed by public charging infrastructures [30]. Additional charging infrastructures with easy availability and high compatibility are linked to reduction in range anxiety and ultimately higher purchase of EVs [3, 7, 29, 31]. Though higher range EV is an alternative, whether it covers all desirable trips without long recharging hours and associated additional infrastructures are uncertain [13].

To elucidate the standard levels of charging, the first one is level 1 charging or slow charging or home charging which uses an AC connection of 110 V. With this type of charging, it takes 8–12 hours for an EV like Nissan Leaf (160 km range) to be fully charged. The second one is level 2 charging which uses an AC connection of 220 V. With this type of charging, it takes half the time compared to level 1 charging for the same model to be fully charged. The third one is level 3 charging or DC charging which uses DC connection of 480 V or higher. With this type of charging, it takes 15 minutes for the same model to be fully charged [5].

4.1.1 Increasing the number of charging infrastructures

While home-based chargers are sufficient in the initial stage, especially for those with low daily requirements, those who make frequent long trips and intermediate stops get concerned about public charging facilities [31, 32]. It is important to embed such scenarios in analyzing range anxiety to get a realistic reflection [33]. On a different basis for reasoning, as the countries are binding themselves legally to reduce carbon footprint, they are aiming for massive replacement of the existing vehicles by EVs. A study conducted in UK concludes that for 100% replacement of ICEVs by EVs, charging infrastructure has to be increased [34]. In case of market share of EVs in Michigan, another research has pointed out the need to invest significantly more on charging stations for intercity trips [35]. In case of EVs with battery capacity above 60 kWh, accessibility of charger upon reaching a range of 25 km reduces delay in the journey and prevents range anxiety [32]. Enabling more charging locations also helps to deal with battery capacity fade over time and prolongs useful life of batteries [36].

4.1.2 Optimizing the location of charging infrastructures

Mass adoption of EVs is dependent on sufficient recharging network optimally located based on customer behavior and psychology eventually focusing on range anxiety [37, 38]. Due to narrow range of EVs, travel behavior is different compared to that for ICEVs. EV users take into account charging time and distance from the origin, and are moreover consistent with travel direction [39]. A model has been developed for Italian highways that creates a map considering road system and availability of infrastructures beside drivers’ behavior concerning range anxiety [40]. Analogous model developed for Beijing suggests installment of 200 public charging stations and 70% home chargers to fulfil 90% of travel demand without changing their daily trips [15]. Even distribution of optimal number of fast charging stations facilitating larger group of travelers is achieved using Geographic Information System-based location optimization method in Hungary [6]. A supplementary feature allowing for smart route selection so as to avoid congestion is also studied in large-scale transportation network [41]. A study has developed such models for fast and slow charging, while minimizing total cost, for urban areas [42]. Such limited budget optimization model has also been developed while curtailing computation time using a compact mixed-integer nonlinear programming [43]. It should be noted that locating charging stations for inter-city and intra-city networks require different approaches. For highways, the existing refueling and resting places are suitable whereas for urban areas, the areas where vehicles are frequently parked for a longer term serves the purpose [44]. For such parking lot, an octopus charger in the middle can reduce inconvenience and avoid delays [7]. In addition to installing strategically planned number of charging stations, it is also important to use a communication protocol among EV service providers so that users can use facilities of any network [45]. Internet of Things (IoT) has been used in a real-time forecasting application to avoid waiting time while securing privacy [46].

4.1.3 Workplace charging

Workplace charging is gaining attention because for those places having home charging facilities in place, introduction of charging facility in workplace would reduce failure rates by more than 25% because breaking down the home-to-home tours decreases the distance between two charging opportunities [47]. Preferable distance between two charging stations is 7 km. But this value can vary depending on whether the area is rural or urban [18]. For high mileage commuters workplace charging increases utility from 57% to as high as 75% [3]. Nevertheless, the employers need to be supported via subsidies and electricity tariff adjustment to promote workplace charging provision [48].

4.1.4 Upgrading charging infrastructures

Increasing the number of charging stations and locating them optimally still may not fix range anxiety problem in relation to charging infrastructures entirely. Efficiency and level of charging should also be considered [49]. A 400 kW charger allowing 200 miles in a time of 10 minutes can match with the 5 minute refueling time of an ICEV [50]. A case of Finland and Switzerland has shown that fast charging stations are necessary to maximize the potential [13]. While increasing level of charging at home-based facility (level 1) has negligible impact, widely installed level 2 charging in public places can help achieve 100% of the low mileage requirements and greatly handle high mileage concerns [3]. Delay in charging process can further be ameliorated via DC fast charging infrastructure [51]. In case of inter-city travel, it has shown to alleviate range anxiety cost-effectively in case of California [52].

Following a mass adoption of EVs, a momentous capital expenditure in fast-charging is inevitable [32]. A study, however, suggests that level of charger power does not necessarily affect adoption rate of EVs in relation to range anxiety [47]. Instead adding the charging facilities that can easily be utilized by EV users has larger implication because EVs are parked for a large amount of time [11]. Also, fast-charging comes together with high electricity cost [51]. It calls for a mix of cheaper slower and expensive faster charging infrastructures. Presence of just a single fast-charging facility has shown to remarkably lower range anxiety but unplanned excessive number of installations would amass immense costs [17]. Urban areas can have access to several service points where drivers can stay long enough to use a level 1 charger [44]. Inter-city travel may demand fast chargers, but considering a low budget scenario, level 1 chargers are preferable [38]. The profits linked to user satisfaction concerning range anxiety has to be compromised with the cost of upgrading charging facilities [53].

4.2 Estimating accurate range

Not arriving at the destination on time induces range anxiety in EV drivers. The range prediction system has to take real-time factors into account for it to be accurate [26]. It has to integrate the effect of slope, acceleration and driving behavior [25]. Machine learning-based framework [54], multilevel mixed-effects linear regression models [55], stochastic battery depletion models [56], deep convolutional neural network [57], radial basis function neural network [58] and Markov model [59] are used in studies to approximate real-world energy consumption better. Use of sensors required to record real-world data can be avoided by using parameters such as vehicle speed, tractive force and gradient [57]. Considering the effect of ambient temperature in non-linear models reflects the environment better [28]. Details of the planned route can also be analyzed to prepare appropriate algorithms such as Adaptive Neuro Fuzzy Inference System, Model Predictive Control and Least Mean Squares [60]. Further, effect of the decay of battery packs (60% of the normal in 5 years) on the delivered range should also be accommodated [58]. Enhancing precision of range estimator ultimately helps in future planning and sustainable management of charging infrastructures [26, 54].

4.3 Optimizing energy consumption

Driving range of an EV can be optimized based on the working temperature, driving speed, gearbox selection and charging habits. Referring to the first case, the energy consumption can increase by 81% due to ambient temperature [27]. The most economical range of temperature in terms of energy usage for an EV is 21.8–25.2°C. Rational use of auxiliary loads, whose operation is dependent on ambient temperature, can save a mean of 9.66% electricity per kilometer [26]. Increasing accessible fast-charging stations has the ability to reduce impact of temperature but at the cost of battery life. Further, EVs with longer range would ameliorate the situation, but this would require larger batteries [27]. Referring to the second case, for moderate speed, the energy consumed per unit distance travelled is lower compared to that at higher speed [61]. In a study conducted in Beijing, the optimal driving speed is lower (48.97 km/h) for low temperature and higher (51.37 km/h) for high temperature [28]. A cruise control method takes into account the traffic status ahead of the journey and controls vehicle speed to reduce average energy consumption by 23.56% by preventing unnecessary braking [62]. Referring to the third case, selection of automatic gearbox, its design and control strategies are also important in maximizing drivable range [63]. Finally, referring to the fourth case, charging behavior using ‘as-late-as-possible’ approach with a buffer between 5% and 60%, preferably 30%, can prolong battery life by a factor of 2 or more [64]. Also, higher standby energy consumption of charging station increases the total energy required to drive EVs [63]. Aside from all of these factors that needs to be optimized, the energy losses created due to variable load on the grid while charging and discharging EVs can also be minimized via smart grid [65].

4.4 Range extender

Range extender is yet another approach to calming range anxiety while reducing the requirement of larger batteries and public chargers via on-board electricity generation [66]. A gasoline based engine used for the purpose with high compression ratio, exhaust gas recirculation, Atkinson cycle and single-point optimization of all parameters has shown to reach a maximum efficiency of 40.2% [66]. BMW i3 provides an emergency engine at an extra cost of about $5000 in Sweden [67]. These options however compromise with environmental benefit. So, another research recommends use of fuel cell range extender due to its zero-CO2 emission benefit. The research endorses use of fuzzy charging strategy to enhance battery’s state of charge, lifetime and energy efficiency, and make an urban driving experience same as that while using an ICEV [68]. In a case of California, a 40-mile battery EV can be recharged at either home-based charging station or wherever public charging stations are available to conveniently fulfil daily requirements. A range extender powered by a fuel cell provides the user with a total range of over 300 miles [69]. But, this type of range extender is only suitable where Well-to-Wheel (WTW) Green House Gas (GHG) emission is low for hydrogen production alike California.

4.5 Car sharing

Car sharing accentuated by mobile computing technology can be an accessible and affordable option [53]. It can either be a one-way system [70] or a fleet operation [71]. The latter one can offer more benefit. The fleet size has to be designed based on battery recharge time and vehicle range. With level 2 charging facility, an autonomous EV with a range of 80 miles can replace 3.7 private vehicles and that with a range of 200 miles can replace 5.5 private vehicles in case of Austin, Texas. And, with level 3 charging facility the ratio would increase further while lowering the demand on electric grid, but at the cost of higher investment [71].

4.6 Modular battery development

Upgrade in range [8], battery autonomy and reliability in estimation [14] are indeed the core aspects of battery development. But, aside from that, modularity of batteries that adds on portability are seen as cheaper alternatives to tackle range anxiety. A research suggests a strategy that involves charging a battery pack once a day and another battery pack once a week [72]. As of now, lithium ion battery is the one with high energy density [73]. Modularity of EV battery can be attained at an energy density over 0.4592 kWh/kg. Integrating thin film solar power technologies are shown to drop the mark to 0.4083 kWh/kg [72]. Modularity of the batteries is the means to foster battery swapping technology [29]. Electricity consumption associated with battery swapping stations in a network has been determined in a study [74]. The batteries should also be made interoperable across different automobile markets, brands and manufacturers [72]. And, it needs to be stressed that such swapping stations profitable only for the paths with more potential EV customers [75].

4.7 Vehicle solar roof installations

Vehicle solar roof installations allow for recharge of batteries utilizing the same time and space used for parking vehicles. But the maximum benefit of solar add-on can be realized only if the user is dependent solely on home charging. Using workplace or public charging facilities would cause waste of significant fraction of daily insolation. Using level 2 charging at home and workplace prevents capture of 80% of sunlight energy by the system [76].

4.8 Lane expansion

EV charging requires longer time compared to refueling time of conventional vehicles. As a result, the vehicles would have to form a queue in a transportation network if the demand is high. This can cause congestion in the route if the charging station does not have enough space for large amount of waiting vehicles. Identifying such critical sections of the route prone to congestion due to high travel demand and prioritizing them for lane expansion facilitates space for charging and reduces total travel time of EV users. Such lane expansion model can be developed considering charging behavior and uncertainty in travel demand, thus addressing associated range anxiety. A study has built such model and obtained optimal model for specific investment scale decreasing the total travel time by 28.54% [77].

Table 1 shows the extraction chart of the scoping review presented in this paper. Remarks column shows inference of the corresponding article towards measures or perception or factors associated with range anxiety in EV users.

Table 1

Charting the Data

Reference	Objective	Method	Key Findings	Remarks
[3]	● To examine sensitivity of battery EV with range anxiety and different levels of installation of charging infrastructures	● Includes time schedules, power levels and locations for different scenarios of charging facility installation ● Applies battery lifetime analysis tool	● While additional level 1 charging facility with reduces range anxiety, higher level of charging does not increase the utility significantly ● Workplace charging helps assuage those with long commute distance ● Increased charging infrastructures addresses range issue of lower mileage drivers completely and that of higher mileage drivers significantly	● Suggested Measure: Increase in number of level 1 charging stations including stations at workplace
[6]	● To develop GIS based location optimization method for fast charging stations	● Ensures even coverage along the route with minimum number of installations of fast chargers ● Considers traffic volume and population of nearby localities of Hungary ● Applies multi-criteria decision-making method to identify suitable installation sites	● Even distribution of fast charging stations ensures their optimal utilization ● Addresses range requirements of low-range EVs	● Suggested Measure: Location optimization of Level 3 charging stations
[7]	● To increase charger accessibility for EV users	● Considers parking configurations, charger design, parking exclusively for EVs, procedure in charging chronology and probable legislations ● Uses data from academic publications, trade market press, dialogues, observations and existing laws	● A charger configuration, such as octopus type, accessible to EV in every possible direction, realistic charging fees, indication whether an EV is fully charged and use of etiquette cards increase usability of chargers and lessen range anxiety	● Suggested Measure: Accessibility optimization of charging stations
[8]	● To analyze users’ concern about range of EVs	● Uses experimental data on purchase decision in California to study statistical behavior using Bayes estimates ● Considers willingness to pay for marginal range improvements, and buy an EV, compensating variation after improvements and demand of range	● Users have high willingness to pay for EVs with improved batteries offering higher range	● User’s perception: Longer range linked with higher EV purchase decision
[10]	● To study cognitive perception of EV users	● Combines both qualitative and quantitative approaches ● Considers subjective perception to find potential for shift towards e-mobility	● Merely comparing EVs and ICEVs cannot draw complete user perspective ● Prior experience shifts user perception towards EVs in positive direction due to better knowledge regarding range limitations and charging processes	● User’s Perception: Prior experience linked with less range anxiety
[11]	● To study if present day EVs provide necessary range to users in US	● Uses detailed physics-based models of EVs along with daily behavior of users ● Quantifies the sensitivities to terrain, ancillary power consumption and battery degradation	● Daily range is below 100 km most of the time ● Level 1 charging alone suffices 89% demand of users on a weekday and 85% on a weekend; the values reduce to 70% and 74% in case of 3% gradient ● Due to large amount of parking time, increasing the number of accessible charging infrastructures have higher advantage over upgrading their charging rates	● User’s Perception: Daily demand addressed by EVs ● Suggested Measure: Increase in number of lower levels over Level 3 charging stations
[13]	● To assess the effect of limited range	● Constructs a simulation model through surveys representative of travel pattern in the subject areas ● Calculates potential of EVs to cover trips and explores ways to increase coverage in relation to charging infrastructures	● EVs prevalent since 2016 can cover 85–90% of trips ● More charging stations and high-range EVs can increase the trip coverage to 99% ● Fast charging infrastructures maximizes the potential of EVs to cover the trips	● User’s Perception: Daily demand addressed by EVs ● Suggested Measure: Increasing number and upgrading charging stations
[14]	● To study relation of user’s perception on EV and its uptake	● Conducts a three-month long experiment for a sample size of 48 people to capture their perception on EV on a rating scale	● General response towards EV is positive mostly supported by less noise, high acceleration and safety ● Battery autonomy and range estimation have to be improved to increase user satisfaction ● Users tend to expect higher than required range buffer to reduce range anxiety	● Suggested Measure: Accurate range estimation ● User’s Perception: Over-estimation of required range
[15]	● To locate public charging stations based on driver’s activities	● Determines charging choice by examining existing activities, charger availability at home and public locality, range anxiety and energy consumption of remaining trips to simulate driver’s charging choice behavior ● Models a location map for charging stations to maximize existing activities ● Applies the model to high-tech areas of Beijing	● Energy required for 46% of drivers in 5 weekdays exceeds the available range ● Adding charging stations (200 public charging stations in this case) can fulfil 90% of travel demand without any change in daily activities ● Changing the day-to-day activities is mandatory to fulfil entire range demand of all of the drivers even if larger charging network is installed	● Suggested Measure: Increase in number of charging stations ● User’s Perception: Change in travel behavior mandatory
[16]	● To illustrate differences in experiencing range anxiety in EV users in critical range situation	● Conducts a field experiment with 74 participants who drive a 94-km round trip ensuring that it reaches critical range situation	● Being familiar with the route, trust in the estimated range, knowledge about control and systems, emotional stability reduces range anxiety	● User’s Perception: Familiarity with EV technology and route preferable ● Suggested Measure: Accurate range estimation
[17]	● To study effect of different scenarios of charging infrastructures on range anxiety	● Studies on both fast-and-expensive and slow-and-cheap charging stations ● Utilizes safe range inventory tool to capture different aspects of range safety assessments	● Range anxiety can increase within a narrow band of 15 km remaining range ● The cut-off point for psychological concern has to be addressed while planning charging infrastructure ● Presence of just one fast charging station can reduce range anxiety but only adding more fast chargers can incur high costs	● Suggested Measure: Optimizing location of charging stations and installation of a few Level 3 chargers
[18]	● Compare range anxiety among experienced and non-experienced EV users	● Includes more than 200 participants in a survey ● Gathers perception on relationship between existing infrastructures for ICEV refueling and required ones for EVs, and dependency of range anxiety with SoC and remaining range	● Users share a common belief that infrastructure for refueling in case of ICEV is more developed, and optimal locations for both gas stations and charging stations should be decided ● Users are influenced by their type of settlement: city dwellers opt for shorter distances between charging stations, preferably 7 km in the region under study ● Experienced EV users are less sensitive to range anxiety due to their prior experience in real environment with the vehicle	● Suggested Measure: Optimizing location of charging stations ● User’s Perception: Prior experience linked with less range anxiety
[19]	● To find out if range related knowledge and prior experience have influence on range anxiety	● Conducts field experiment with 63 participants (experienced and inexperienced) ● Emulates critical range situation	● Range related knowledge and prior experience reduces range anxiety ● Providing relevant knowledge and training before purchasing EV and during usage phase reduces range anxiety	● User’s Perception: Prior experience linked with less range anxiety
[20]	● To study adaptation levels of non-experienced EV users	● Analyses if prior experience on EV driving has any effect on critical range situation ● Conducts field experiment of 74 participants with 94 km unaccompanied round trip ensuring small range safety buffer	● Range anxiety is higher at the start of the trip compared to the end of the trip ● Providing coping information can positively affect range related concerns	● User’s Perception: Prior experience linked with less range anxiety
[21]	● To study effect of in-built information system of EV on range anxiety	● Conducts a field experiment with real traffic conditions ● Emulates critical range situation to analyze psychological response	● In-built information system reduces range anxiety in EV users; range anxiety is more driven by negative attitude towards using EV ● Psychological indicators helped questionnaire evaluation	● User’s Perception: Prior experience linked with less range anxiety
[22]	● To create typology of EV driving practices qualitatively	● Conducts semi-structured interviews to draw qualitative information ● Synthesizes decisions and actions of users before, during and after trips ● Includes household routine aspects, travel habits and behavior	● Illustrates EV driving habits in relation to day-to-day routine, travel behavior, digital device usage, etc. to ensure EV uptake while addressing range anxiety	● User’s Perception: Incorporation of daily routine linked with less range anxiety
[23]	● To put forward reasons of failure of an EV company named ‘Better Place’ which had proposed a business model to reduce the cost of batteries and range anxiety	● Explores operational details of the company in Denmark and Israel considering social, technical, political and environmental factors ● Discusses reasons of failure of the company and relates with energy planning, policy and analysis	● Lack of environmental concern among stakeholders, resistance to change in behavior and lifestyle, mismanagement and high capital costs are the main reasons behind the failure of the EV company	● User’s Perception: Change in travel behavior mandatory
[24]	● To examine range anxiety through Hirschman’s Rhetoric of Reaction	● Uses qualitative method to understand role of range anxiety under different perspectives ● Conducts 227 semi-structures interviews with experts at 201 institutions and a survey with about 5000 participants in five Nordic countries	● Policy and investment on public charging infrastructure may fail to address actual concerns of consumers regarding range anxiety ● Expensive charging infrastructure may introduce charger anxiety in users rather than promoting EV diffusion ● Conducting informative programs and non-monetary policies could change response of EV users by reducing range anxiety indirectly	● User’s Perception: Expensive infrastructures linked with more range anxiety
[25]	● To estimate energy by developing a dynamic range estimator	● Incorporates slope, acceleration and driving behavior ● Utilizes response to error in speed and delay in time between throttle to obtain driving behavior	● Driving uphill, accelerating faster, stop-and-go driving cycle and auxiliary loads consumes the most energy	● Factors Affecting Range: Gradient, nature of traffic and ambient temperature
[26]	● To develop energy consumption model and explore interactive effects of ambient temperature and auxiliary loads	● Performs study based on observations taken from 68 EVs in Japan for a year using Global Positioning System (GPS) ● Calibrates using ordinary least squares regression and multilevel mixed effects linear regression	● Not considering ambient temperature that affects use of auxiliary loads leads to exaggeration of energy consumption in summer and underestimation in winter ● The most economic energy efficiency is seen in the range of 21.8–25.2°C.	● Factors Affecting Range: Ambient temperature
[27]	● To assess the effect of ambient temperature on energy consumption and associated route choice and fleet composition	● Formulates mathematical programming ● Performs computational analysis with the help of data obtained from literatures	● Effect of ambient temperature has to be accounted for while making route choice for efficient operation ● Increase in temperature causes increase in total energy consumption, number of customers at charging spots and number of vehicles required for the same fleet ● Increasing the number of charging infrastructures, specifically fast charging ones for the fleet, addresses impact of high ambient temperature but at the cost of battery life ● Using long-range EVs also addresses the issue but at the cost of larger sized batteries	● Factors Affecting Range: Ambient temperature ● Suggested Measure: Increasing number of Level 3 charging stations and long-range EVs
[28]	● To estimate remaining driving range for real-world case	● Collects data from an EV operating in China for 3 months with significant differences in temperature ● Considers State of Charge (SoC), speed and temperature and non-linear relationship between speed and driving distance per SoC	● Economical driving speeds under low, moderate and high temperate conditions are 48.97 km/h, 50.89 km/h and 51.37 km/h, respectively, showing a positive relationship	● Factors Affecting Range: Ambient temperature ● Suggested Measure: Optimizing energy consumption
[29]	● To identify factors affecting range and range anxiety, and Suggested Measure to address them	● Reviews EV trends in global and Indian context ● Focusses on entire EV ecosystem	● Reducing gross vehicle weight, rolling resistance and drag coefficient, optimizing vehicle performance, improving specific energy density and volumetric efficiency, lowering weight per kWh, managing fleet efficiently, carrying out operation research on charging, increasing charging infrastructure, standardizing charging protocol and establishing battery swapping business model can overall increase EV range and reduce range anxiety	● Factors Affecting Range: Inbuilt vehicle/battery characteristics and rolling resistance ● Suggested Measure: Increasing number of charging stations and battery swapping technology
[30]	● To quantify value of chargers as replacement of gasoline for plug-in hybrid EVs	● Simulates value of additional miles offered by public chargers based on offered range, annual miles traveled, existing charging stations, energy prices, efficiency of vehicle and income status of users ● Conducts case study on California’s public charging network	● Public charging stations provide more access and mobility to EV owners, thereby increasing their value ● Public charging stations can take over the additional cost incurred by EV users due to limited range and long recharging time	● Suggested Measure: Increasing number of public charging stations
[31]	● To develop a modeling tool to incorporate traffic flow patterns to address range anxiety	● Considers trip chain as basic unit that governs decision regarding travel route and location choices ● Develops cascading labeling algorithm for shortest path problem	● Range anxiety affects travel behavior if the user makes a series of trips rather than a single trip given the distance between charging stations	● User’s Perception: Number of intermediate stops linked with range anxiety ● Suggested Measure: Incorporation of trip nature for charging infrastructure planning
[32]	● To quantitatively analyze inconvenience of EV charging in relation to ICEV refueling	● Analyzes 39 020 travel diaries of UK and quantifies inconvenience in terms of delay for different combinations of battery capacity, charger capacity and accessibility	● 95% of users with home charging can reach convenience parity with affordable EV model; it is not true for those who rely on workplace or public charging alone ● Delay is negligible if charging facility is available upon reaching a minimum of 25 km for battery capacity higher than 60 kWh ● Development of extensive charging infrastructure is also favorable despite significant expenditure for charging station and grid reinforcements	● User’s Perception: Daily demand met by Level 1 charging ● Suggested Measure: Increasing number of charging stations
[33]	● To develop analytical evaluation tool for EV traffic network	● Considers trip chain level scenario to better represent range anxiety ● Characterizes equilibrium conditions for discrete and continuous driving range distribution ● Uses projected gradient method	● The proposed method addresses heterogeneity of range anxiety in EV networks ● The method incorporates effect of existing infrastructures	● Suggested Measure: Incorporation of trip nature for charging infrastructure planning
[34]	● To investigate users’ perspective on uptake of EVs	● Combines a system integrating transportation, energy and environment with demand and supply chain of EVs ● Extrapolates low carbon scenarios for different investments and policies ● Includes participants who either owns or drives an EV	● Dynamics of EV market during adoption phase has to be incorporated for more realistic future market prediction, evaluate effectiveness of policy changes and assess social issues such as range anxiety ● For the case considered in the paper (for UK), more charging infrastructures and regulations to phase out non EVs are required to ensure 100% replacement of conventional vehicles	● User’s Perception: Incorporation of range anxiety issues important for EV infrastructure planning and development
[35]	● To find a configuration of charging stations for intercity trips	● Ensures minimization of investment cost to build stations and delays including charging time, queue and detour ● Captures realistic patterns of travel	● Significant investment is needed regardless of initial EV adoption level to support intercity trips so as to minimize range anxiety among users	● Suggested Measure: Increasing number of charging stations for intercity trips
[36]	● To find if daily range requirement of EVs are satisfied under battery depletion and power loss	● Applies detailed physics-based models of EVs and uses data of US drivers regarding their driving behavior ● Presents results as how much part of the battery would be depleted at different levels of capacity fade	● EV batteries meet the daily demand even at 80% remaining storage capacity ● Installing at least level 1 charging facilities in more locations lengthens useful life of batteries ● Even at 30% remaining power capacity, EV performance is not affected significantly ● The effect of battery degradation intensifies in case of high ancillary power consumption and uphill driving, which can be mitigated by installing more charging facilities	● User’s Perception: Daily demand addressed by EVs ● Suggested Measure: Increasing number of charging stations to address battery degradation
[37]	● To study charging station placement issue with reference to range anxiety and path deviations	● Formulates problem as a function of range anxiety ● Solves problem using k-shortest path algorithm and iterative greedy heuristic ● Analyzes effects of parameters on real-world scenario	● Battery range should be considered while planning charging infrastructures ● Lower range is associated with higher construction costs and path deviation	● Suggested Measure: Increasing number of charging stations to address limited range
[38]	● To study effect of different levels of public charging on range anxiety	● Uses GPS data for greater Seattle metropolitan area ● Uses optimization model to place chargers at locations such that the position considers charging behavior of the customers on the route and budget allocated for the deployment ● Assumes that travel patterns remain the same when users switch from ICEVs to EVs	● Most of the trips fall within typical EV range ● Optimal location of public chargers reduces range anxiety ● Level 1 chargers are suitable for small budget scenario whereas ● Level 3 chargers are not preferable due to high cost ● Level 3 chargers are preferable for intercity trips	● Suggested Measure: Increasing number of Level 1 and Level 3 charging stations as per budget and trip nature
[39]	● To study differences between charging and route choice among EV and ICEV users	● Uses multinomial logit-based and nested logit-based models ● Categorizes charging decision as upper level and route choices as lower level in the nested structure ● Conducts numerical analysis to verify effect of range anxiety on decision making process	● Initial SoC affects charging decision and SoC at destination affects route choice ● Charging time and location contributes to decision making process; less charging time and location closer to origin are preferred	● User’s Perception: Route choice and charging behavior influenced by charging status, location and total time ● Suggested Measure: Optimizing location of charging stations by incorporating parameters dependent on charging behavior
[40]	● To determine optimal position of EV charging stations along a highway	● Uses the most recent database to map infrastructural need and find out EV market share of Italy ● Considers vehicle autonomy, energy of battery pack, technical characteristics provided by manufacturers, road system, prerequisites for charging system and driver behavior, specifically range anxiety	● A map showing candidate points to allocate charging stations has been made for the region under study ● The map can be used for correct planning and positioning of infrastructures to address range related uncertainties among users	● Suggested Measure: Optimizing location of charging stations
[41]	● To develop a pragmatic method to optimally place charging stations	● Develops methodology to find minimal number of charging stations and their locations to ensure even coverage along a network ● Uses iterative approximation of route node coverage problem ● Applied on Sioux-Falls test transportation network and southern part of Sweden ● Considers already existing infrastructures as well	● The proposed method can be easily implemented in computer coding which increases its utility ● Strategic locations are selected without numerical difficulties to meet demand of EV fleet while avoiding congested routes	● Suggested Measure: Optimizing location of charging stations
[42]	● To find optimal locations of charging stations	Includes both fast and slow charging station Optimizes total cost while ensuring maximum coverage	● The proposed model is practical and effective to decide placement of charging stations to make sure that the stations are evenly distributed	● Suggested Measure: Optimizing location of charging stations by considering level for even coverage
[43]	● To find optimal charging stations under given budget to address range anxiety	● Uses a compact mixed-integer non-linear programming model ● Incorporates charging time, cost and possible path deviation ● Assumes range anxiety profile to be nonlinear function	● The proposed solution is applicable to determine optimal charging locations with less computational load	● Suggested Measure: Optimizing location of charging stations
[44]	● To identify location of charging station for long trips	● Considers demography, economy, environment, transportation aspects and existing charging stations	● For urban area, charging stations may be located at spaces where EVs are parked for longer time ● For highways, existing refueling places and rest places can be utilized ● Less number of fast chargers are required in urban areas due to longer accessibility for urban users to normal chargers	● Suggested Measure: Optimizing location and level of charging stations based on type of settlement
[45]	● To assess development of service interfaces in EV charging station network	● Develops a communication protocol to directly communicate among charging networks ● Ensures communication among networks both registered and not registered by users	● Accessible charging stations, more than just availability, help reduce range anxiety ● The proposed protocol provides cross-network charging facilities	● Suggested Measure: Increasing accessibility of charging stations
[46]	● To provide a real-time forecast of appropriate charging station through server to reduce or completely avoid waiting time through the use of Internet of Things	● Uses PHP programming language in Linux UBUNTU 16.04 LTS and processes through a google cloud platform ● Validated through a low-cost test system ● Considers user convenience, electricity prices and willingness of users to pay	● The proposed method avoids external intervention and protects privacy of the user ● The proposed method can be used for users’ convenience as well as prediction of future load for load balancing and to avoid congestion	● Suggested Measure: Increasing accessibility of charging stations for lower congestion and balance between supply and demand
[47]	● To assess impact of workplace charging on adoption of battery EVs and plug-in hybrid EVs	● Uses GPS based data of 143 vehicles operating for 20 days to 18 months Builds home-to-home trips for each vehicle, determines workplace locations through travel behavior, breaks down the whole trip to home-to-work, work-to-home and work-to-work trips ● Considers three levels of charging and price of gasoline for analysis	● Availability of workplace charging reduces range anxiety in battery EV users due to shorter distance between charging points and causes decrease in failure rates during adoption phase ● Upgrading level of charging does not have significant impact on reducing failure rate; level 1 charging is good enough to meet daily requirements	● Suggested Measure: Increase in number of Level 1 and workplace charging
[48]	● To analyze incentives and barriers related to workplace charging through economical approach	● Considers demand and supply of charging at workplaces ● Considers both employees’ and employers’ perspective on economical aspect of charging infrastructures ● Discusses subsidies, charging costs, electricity tariffs and loading technologies related to workplace charging	● While employees opt for availability of charging station at workplace, employers prefer otherwise ● Subsidies directly meant for workplace charging facilities could promote the installation of such infrastructure	● Suggested Measure: Provision of subsidy for installation of workplace charging
[49]	● To study change in market share of passenger vehicles	● Considers ICEVs and EVs, fuel prices, taxes, vehicle prices and recharging concerns ● Simulates future development using agent based computational approach ● Develops a vehicle choice algorithm considering social factors and consumer’s preference for vehicle attributes ● Creates different scenarios and iterates the model to determine market share evolution until 2030 for Iceland	● Under the scenario of high gasoline price and decreasing EV price without tax and with proper charging infrastructure, EV would replace ICEVs completely ● Under the scenario of low gasoline price or combination of medium gasoline price and unchanging EV price, support policies are need to foster EV adoption ● Range anxiety is not mitigated solely by adding charging stations; recharging time has to be reduced and available range has to be increased.	● Suggested Measure: Upgrading charging infrastructures and using long-range EVs
[50]	● To improve recharging rates of EV to make it comparable to ICEV refueling	● Considers vehicle system design and recharge time ● Assesses end impact on system voltage and vehicle components	● Increasing the charging power to at least 400 kW so that the EV provides 200 miles range in 10 minutes recharge time makes it comparable to ICEV refueling time ● Faster recharge is linked to lessened range anxiety in long-distance transportation	● Suggested Measure: Increase in number of Level 3 charging stations
[51]	● To propose DC fast charging as a solution for long trips and explore mitigating measures for associated electricity cost	● Analyzes cost components in US and real-world vehicle charging load scenarios for over 7000 commercial electricity retail rates ● Assesses if use of solar photovoltaics and batteries can reduce the cost of DC fast charging	● DC fast charging stations are expensive with low utilization compared to home and workplace charging ● Utilization of batteries can help reduce cost specifically for low-utilization loads and that of solar photovoltaics help for loads that are more correlated with solar production	● Suggested Measure: Increase in number of Level 1 and workplace charging
[52]	● To plan location of fast charging infrastructures for intercity route	● Uses mixed integer programming model to determine working status of charging stations and number of chargers required in each station	● Intercity charging stations alleviate range anxiety ● DC chargers has to be increased over time to facilitate growing demand	● Suggested Measure: Increase in number of Level 3 charging for intercity routes
[53]	● To assess impact of battery capacity on car-sharing	● Uses discrete event simulation approach Assesses user’s perspective on limited battery capacity	● Faster charging speed, higher range and greater vehicle-to-trip ratio assuages user’s concern regarding limited battery capacity ● Trade-off has to be made for optimal investment, vehicle usage and user satisfaction so as to avoid dependency on fastest chargers and longest range	● User’s Perception: Faster charging and longer range EV linked with lesser concern for limited battery capacity ● Suggested Measure: Upgradation of Level 3 charging stations ensuring optimal investment
[54]	● To estimate actual energy efficiency of EVs	● Extracts impacts of road environments and traffic conditions of real-world driving on energy efficiency ● Tracks GPS data of 68 EVs in Japan for a year ● Compares different types of EV ownership, external environments and driving behaviors	● Different data clustering methods provide different estimates of influencing factors under consideration ● The presented method improves the energy consumption estimation by 7.5%	● Suggested Measure: Accurate range estimation
[55]	● To develop energy consumption prediction framework	● Uses novel machine learning on real-world driving data collected from fifty-five electric taxis in Beijing city ● Uses data fragmentation technique (trip, micro and kinematic) Extracts vehicle, environment and driver-related factors of energy consumption	● The proposed method increases the accuracy of energy consumption prediction by nearly 30% which in turn alleviates range anxiety	● Suggested Measure: Accurate range estimation
[56]	● To assess problem of stochastic battery depletion in EVs in pickup and delivery	● Develops a chance-constrained mixed integer non-linear programming model and performs linear approximation	● Delivery plans should include uncertainties associated with stochastic battery depletion to reduce range anxiety	● Suggested Measure: Accurate range estimation considering battery degradation
[57]	● To accurately estimate real-time energy consumption	● Uses deep convolutional neural network Requires vehicle speed, tractive effort and road elevation ● Explores impact of different parameters and architectures in simulation environment ● Uses Nissan Leaf 2013 model	● The proposed method provides range estimation with less error and thus help reduce range anxiety in EV users	● Suggested Measure: Accurate range estimation
[58]	● To estimate remaining range of EV	● Uses radial basis function neural network method ● Considers non-linear system for battery factors and vehicle factors ● Uses contribution analysis method to enhance estimation method	● Range estimation errors are reduced Battery decay of 60% in 5 years causes rapid decrement in remaining range	● Suggested Measure: Accurate range estimation considering battery degradation
[59]	● To calculate remaining discharge energy of battery	● Uses a stochastic load prediction algorithm through Markov model and Gaussian mixture data clustering ● Validates the results experimentally under real-world dynamic current profiles	● The proposed estimation system predicts future values of the battery load current, terminal voltage, temperature and other parameters and improves accuracy	● Suggested Measure: Accurate range estimation
[60]	● To suggest improvements in range estimation of battery EVs	● Analyzes nature of algorithms for calculation of range ● Uses data from travel routes and creates methods to identify range calculation algorithm ● Evaluates the identified methods	● Improving algorithm in calculating range improves accuracy and ultimately reduces range anxiety	● Suggested Measure: Accurate range estimation
[61]	● To study routing choice of EV and ICEVS users, and its effect on transportation network	● Determines equilibrium flows for traffic consisting of both EVs and ICEVs ● Conducts simulation to find solution ● Selects route based on minimal cost associated with travel time, energy and range anxiety	● EV users prefer lower speed routes to conserve energy and reduce range anxiety, whereas ICEV users prefer shortest routes unless the route is congested	● Suggested Measure: Optimizing energy consumption by selecting suitable speed
[62]	● To improve range of EV through cruise control	● Controls speed depending upon upcoming traffic signal so as to maintain a certain distance with preceding vehicle ● Studies a Tesla S model in simulation environment with different initial conditions, and in presence and absence of preceding vehicle	● Cruise control method improves EV’s energy efficiency and thus range ● Energy consumption is reduced by approximately 23.56%	● Suggested Measure: Optimizing energy consumption by selecting suitable speed
[63]	● To analyze energy efficiency and range of EVs with change in gear selection	● Uses EV-converted vehicles on a chassis dynamometer to control the environment and comply with international standards ● Uses two models: with automatic gear drive and with manual gear drive ● Measures electricity consumption under different gear setup for identical driving cycles	● Testing on chassis dynamometer overestimates energy consumption due to relatively cold drivetrain ● Standby energy consumption has to be reduced for less energy requirement ● Gearbox selection, design and control for automatic type plays an important role in energy consumption, however using different gears for manual type	● Factors Affecting Range: Gearbox selection
[64]	● To optimize charging according to users’ requirement to prolong battery life	● Implies comprehensive battery aging model to find optimal charging ● Considers charging while in an SoC range of 5–60% (as-late-as-possible charging approach) ● Analyzes battery life while considering range flexibility	● Optimal charging doubles the expected life of battery ● While customers tend to full charge as soon as charging facility is available, it compromises with battery life	● Suggested Measure: Optimizing energy consumption with optimal charging of battery
[65]	● To minimize operational cost and energy losses through smart grids	● Considers adjustable loads for energy scheduling because to intensified effects in power losses associated with vehicle-to-grid (V2G) system ● Formulates and optimizes range anxiety to address battery depletion issue and encourage V2G system	● Battery depletion has high influence on V2G contribution of EVs ● Higher range anxiety causes users to ensure higher SoC and in turn overloads the grid ● Trade-off is required between range anxiety and V2G service	● Suggested Measure: Optimizing energy consumption with incorporation of smart grids
[66]	● To investigate use of a 3-cylinder gasoline spark-ignition engine as range extender	● Applies Atkinson cycle, circulates exhaust gas and uses direct injection of gasoline to suppress knocking ● Optimizes around a single operating point by combining artificial neural network and genetic algorithm	● Thermal efficiency of the range extender is higher at lower power requirement; enhancing efficiency of range extender serves to reduce range anxiety	● Suggested Measure: Using range extender
[67]	● To compare plug-in hybrid EV with battery EV in terms of economic viability and replacement potential	● Considers two-car households for the study to ensure that EVs can provide benefit optimally ● Uses GPS data from 64 households of Sweden	● Provided that battery EVs are optimally utilized, mass produced, battery costs reduced and home charging option is only available, battery EV has lesser total cost of ownership and is more preferable than plug-in hybrid EVs ● Additional engine can be used alongside battery EV to cope with range anxiety	● Suggested Measure: Using range extender
[68]	● To present a fuzzy charging strategy for power decentralized fuel cell and battery EVs	● Includes smooth steering experience and range extension based on energy management in urban driving ● Uses simulation environment for analysis of feasibility and effectiveness	● The proposed strategy enhances SoC, improves range, makes urban driving of EV and ICEV comparable	● Suggested Measure: Using range extender
[69]	● To analyze fuel cell EVs	● Considers range provided by battery incorporated in the fuel cell EV along with that provided by hydrogen fuel ● Uses California as a case study to analyses if the daily demand is within the provided range	● 40-mile range provided by battery covers most trips which can be replenished through home charging or public charging wherever available ● Hydrogen fuel requirement decreases significantly than that of pro	● Suggested Measure: Using range extender
[70]	● To develop a practical one-way char sharing system	● Uses mixed integer programming ● Considers charging demand, charging modes and range anxiety	● User’s preference affects depots location and fleet size of sharing system	● Suggested Measure: Car sharing
[71]	● To manage shared fleet of EVs for Austin, Texas	● Considers available charging infrastructures and range of market available EVs ● Uses agent based discrete-time model	● Fleet size depends upon recharge time and range of available EVs, shorter recharge time and longer available range can replace relatively more privately-owned vehicles ● For the case considered, 80-mile range EV fleet with level 2 charging stations require the least investment for operation, but requires large parking areas to flatten of peak demand; the peak demand is addressed through the use of level 3 charging but with higher cost per mile	● Suggested Measure: Car sharing
[72]	● To study whether portable battery may be used for EVs	● Optimally sizes modular batteries, uses thin film photovoltaic cells in windows and optimizes charging strategy ● Analyzes charging convenience and carrying battery module to derive optimal size	● The paper endorses use of a daily charged battery and a weekly charged battery simultaneously, that is standardized and interoperable across different manufacturers, to allow for the use of smaller batteries ● Portability of EV battery become feasible beyond storage density of 0.4592 kWh/kg or use of thin film solar technologies which reduces the mark to 0.4083 kWh/kg ● Portable batteries reduce need for extensive charging infrastructures	● Suggested Measure: Developing modular battery
[73]	● To investigate relationship between different models of battery and range anxiety	● Considers Lead acid, NiCd, NiMH and Li-ion batteries ● Simulates the batteries in a common EV model and compared distance travelled	● Li-ion provides more travel range due to its high energy density and thus addresses range anxiety better	● Suggested Measure: Using Li-ion battery due to higher energy density
[74]	● To address traffic congestion and electricity consumption issues while charging EVs in battery swapping stations	● Considers electricity consumption on a route due to flow of EVs ● Describes route choice behaviors of EV users Presents models and solution algorithms to formulate user equilibrium conditions ● Provides evaluation and improvement methods Provides congestion pricing model to minimize travel cost	● Flow-dependent electricity consumption has high influence on route choice behaviors of EV users	● Suggested Measure: Battery swapping technology
[75]	● To design network for swapping and charging stations considering battery leasing and car-sharing business models	● Relates user satisfaction with remaining range ● Formulates problem as linear integer programming model ● Conducts parametric analysis on real-world scenario	● The business models are preferable due to affordability ● The business models for battery leasing and car-sharing are more preferred in routes with more users inclined towards EVs	● Suggested Measure: Battery swapping technology and car sharing
[76]	● To examine relevance of vehicle solar roof and workplace charging among EV users	● Emphasizes on users who have access to workplace charging ● Examines if solar installation on the roof is worthwhile during parking where charging station is available ● Compares the utility with those having level 1 charging at home as well	● Level 2 workplace charging station causes loss of over 75% solar energy that could be utilized from solar roof installation ● Charging while plugged in during parking at workplace fills up most of the storage leaving less room for solar energy storage ● For users who charge at home as well, the loss percent of potential solar energy storage increases to 80% ● In absence of workplace charging, solar roof installation is found to have the most utility	● Suggested Measure: Installation of workplace charging
[77]	● To design optimal lane expansion for EV transportation network by minimizing total travel time	● Designs local optimal solution algorithm to create network design model ● Considers charging behavior and range anxiety in EV users ● Uses robust optimization model to reduce uncertainties in transportation demand Performs sensitivity analysis of control parameters and government investment scales	● The proposed scheme reduces total travel time by 28.54% ● Critical links in the network should be focused to reduce travel time ● Government should determine investment scale carefully for lane expansion based on the effects of initial implementation	● Suggested Measure: Lane expansion for critical sections of a route considering investment
[78]	● To compare fuel economy of battery EV, plug-in hybrid EV and hybrid EV with ICEVs, and relate range anxiety with associated prices	● Uses data on new cars sold in 8 EU countries for over 6 years, and information about gasoline, diesel and electricity prices and taxes	● Higher fuel efficiency is followed by higher prices because of either undercapitalization of fuel economy or shorter expected payback period ● Prices of EVs are not reflected in their battery range	● User’s Perception: Battery range not in agreement with EV price

5 DISCUSSION

Although current days EVs are shown to have the capability to satisfy daily travel needs, shorter range and longer recharging time in comparison with ICEVs along with occasional failure in meeting the travel demand have negatively impacted on the attitude of users towards EVs. While the environmental condition, geographical status and local driving circumstances determine the range an EV can offer, additional strategy would help resolve range anxiety among EV users and cope with the distinctive aspect of the technology. Range anxiety, though being a psychological problem, requires technical solution in the long run. This study has identified eight measures via scoping review of articles which directly and indirectly addresses the factors that induce range anxiety as shown in Table 2.

Table 2

Range Anxiety Factors Addressed by Mitigating Measures Listed in this Study (‘✓’ represents ‘directly addresses’ and ‘×’ represents ‘indirectly addresses’)

Addressed Factors	Unfamiliarity with the technology	Stress due to limited range	Low route familiarity	Distrust on range estimation	Change in daily routine	Change in parking practice
Measures ↓
Increase the number of charging stations	|$\times$|	✓		|$\times$|	✓
Optimizing the location of charging infrastructures	|$\times$|	✓	✓	|$\times$|	✓	✓
Installing workplace charging	|$\times$|	✓		|$\times$|	✓	✓
Upgrading charging infrastructures	|$\times$|	✓		|$\times$|	✓
Estimating accurate range	|$\times$|	✓	✓	✓	|$\times$|
Optimizing energy consumption	|$\times$|	✓	|$\times$|	|$\times$|	✓
Using range extender	|$\times$|	✓	✓	✓	✓
Car sharing	|$\times$|	✓			|$\times$|
Developing modular battery	|$\times$|	✓	✓	|$\times$|		✓
Installing solar roof on vehicles	|$\times$|	✓	✓	|$\times$|	✓	✓
Expanding lane	|$\times$|					✓

All the measures indicated in this paper has to be implemented chronologically. A strategical plan has to be followed as shown in Figure 2. Key to resolving range anxiety differs in terms of level of EV diffusion in the market [9]. Estimating accurate range and optimizing energy consumption are essential regardless of level of diffusion. EV drivers need to have precise idea about the remaining distance they can travel given the driving conditions at hand [60]. Distinct geography, climatic zones, traffic status, social context and driver behavior demands for a local driving cycle that helps in estimating remaining range correctly and would not be the same for all of the regions. They eventually reduce stress inducing factors in EV users.

Figure 2

Stage Relation among the Listed Mitigating Measures

At an early stage, home charging would suffice the daily requirements of EV users. Investment on increasing number of charging stations anywhere or upgrading them would only strain the economy of a country [53]. For drivers with longer range requirement, in which case they intend to have access to charging at convenient locations [32], workplace charging may be introduced. This can counter the range anxiety associated with commuting distance allowing the users to not change their daily travel routine. With increase in number of EVs plying on the road, public charging stations have to be built. But rather than just installing them haphazardly, optimal location should be identified for the routes where concentration of EVs is high. This helps in uniform and sparse distribution of such stations so as to attain maximum gain under minimum investment [6]. Appropriate mapping of these stations would take care of route unfamiliarity among users.

Higher level of charging such as DC charging would still not be appropriate at this stage from economical viewpoint [17]. Range extender can be used to travel an extra mile. Apparently, it seems to provide a solution to all of the stress-inducing factors. For longer range requirements, car sharing can also come into picture. If higher budget were to be allocated in vehicle development itself, solar roof installations would also help achieve the target range. But, this should not go simultaneously with frequent charging from workplace or public stations [76]. With increase in charging stations, congestion that may be caused due to longer recharging time of EVs should also be handled. Along with optimal location of charging stations in a certain route [41], lane expansion helps in prevention of such blockage [77]. Thereupon DC charging facilities should also be introduced gradually at optimal locations [32]. With larger number of charging stations to choose from, interoperability among service providers should also be ensured [45]. This further boosts introduction of modular batteries that can be easily transported to nearby charging locations [72].

6 CONCLUSION

Range anxiety has been identified as a negative influence for sustainable diffusion of EVs in multiple literatures. Prompting vast majority of customers to purchase EVs requires long term solution to cope with limited range of EVs. This paper has categorized studies focused on range anxiety in EV users into user’s perception, factors affecting the range and mitigating measures after a systematic review of literatures. Although present-day EVs can satisfy daily requirements, users are doubtful regarding the same. Real world factors affecting the range of an EV, such as traffic, gradient, ambient temperature, driving speed, etc. have been recognized and being worked on. These factors have to be incorporated in determining accurate range of EVs regardless of level of diffusion. Aside from this, optimization of energy consumption, optimal and even distribution of charging stations, use of range extender, introduction of car sharing, development of modular batteries and lane expansion helps to assuage the anxiety inducing elements in users. An alternative to additional infrastructures is installation of solar roof so as to exploit daily insolation to the greatest degree possible. These measures should be implemented by policy makers and stakeholders systematically starting from precise range estimation and optimal energy consumption strategies to adding and developing infrastructures and apparatus. Further studies can be done on devising detailed approach to integrate each of these identified measures depending upon the particulars of the region under study.

FUNDING

University Grants Commission Nepal. Grant Number: MRS-7778-Engg-04.