Vehicle Design Strategies to Reduce the Risk of COVID-19 Transmission in Shared and Pooled Travel: Inventory, Typology, and Considerations for Research and Implementation

Abstract

The global COVID-19 pandemic has given rise to a plethora of ideas for modifying and redesigning public transportation and shared mobility vehicles to protect workers and riders from contracting the disease while traveling. This research seeks to inventory these strategies, and to organize and distill them in a way that enables researchers, policymakers, and public transport and mobility service operators to more systematically and efficiently evaluate them. Through literature search and analysis, the COVID-19 risk-mitigating vehicle design (CRVD) typology was developed, articulating 12 categories of strategies (e.g., Seating Configuration, Barriers) and 12 mechanisms (e.g., physical distancing, physical separation) by which the strategies may reduce COVID-19 spread. A secondary contribution of this research is to gather opinions of experts in fields related to COVID-19 and its transmission, about the identified CRVD strategies and mitigation mechanisms. The typology and expert opinions serve as a launching point for further innovation and research to evaluate the effectiveness of CRVD strategies and their relationship to user preferences and travel behavior, within and beyond the current context. Public transport and shared mobility service operators can use the CRVD typology as a reference, in conjunction with industry guidance and emerging research on strategy effectiveness, to aid decision-making in their continued response to the pandemic as well as for future planning.

The COVID-19 pandemic has had dramatic impacts on transportation globally ( 1 ). Widespread shelter-in-place mandates and public health recommendations including social distancing resulted in reduced travel and travel mode shifts—away from shared and pooled travel modes. Transportation Network Companies such as Uber and Lyft halted their pooled ride-hailing options ( 2 ). U.S. public transportation ridership decreased by 79% at the beginning of the pandemic, was down 65% from June 2020 through December 2020 ( 3 ), and remains up to 50% below baseline (depending on the mode [ 4 ]).

Shared and pooled travel modes are critical components of a decarbonized and equitable mobility future ( 5 ). Public transit, carpooling, electric car-sharing, pooled ride-hailing, and micromobility are less energy- and emissions-intensive alternatives to the more dominant mode of single-occupancy vehicles ( 6 ). These modes already comprised a relatively small fraction of travel in the U.S. before the pandemic and now will likely remain further suppressed in the wake of the pandemic if people continue new mode-choice habits and altered perceptions of travel endure ( 7 , 8 ). On the other hand, those who continue to rely on public transportation, including low-income communities ( 9 ), essential workers ( 10 ), and communities with higher proportions of African American, Hispanic, Female, and over-45-year-old residents ( 11 ), are disproportionately at risk to the degree that these modes leave them susceptible to disease transmission ( 12 – 16 ).

For pooled and shared travel to safely return to and ideally surpass pre-pandemic levels, it is important to implement solutions to reduce the real and perceived risks of infectious disease transmission ( 10 ). Solutions may involve new policies and business models, public awareness programs, and innovative station and vehicle design. This research focuses on vehicle design strategies to facilitate safe and confident use of shared and pooled travel modes in the wake of the pandemic.

Industry Guidance

Pooled- and shared-travel service operators have been given guidance on vehicle design strategies to reduce the risk of COVID-19 spread, but there is a need for a more systematic approach. Most official guidance is industry specific. For example, the Centers for Disease Control and Prevention (CDC) has provided separate guidance for different types of transit, including bus ( 17 ), rail ( 18 ), and rideshare/taxis ( 19 ). The American Public Transportation Association (APTA) has adapted CDC guidance for public transportation (bus and rail) ( 20 ). Other examples include the “Runway to Recovery” ( 21 ) for mitigating risk in air travel from the U.S. Department of Transportation; guidance for mass transit and marine operators on heating, ventilation, and air conditioning (HVAC) ( 22 ) from the American Society of Heating, Refrigerating and Air-Conditioning Engineers; and recommendations for protecting farm workers in employee vehicles ( 23 ) from the Occupational Safety and Health Administration (OSHA). The California Department of Public Health (CDPH), State Transportation Agency (CalSTA), and Cal/OSHA issued industry guidance that was unique in covering both public and private passenger carrier services ( 24 ). Many similar design strategies (e.g., barriers and seating reconfiguration) are recommended across shared and pooled vehicle and service types. Resources tailored to each industry are important, but analyzing applications across public transit, shared mobility, and air travel could yield a more comprehensive and nuanced shared understanding of vehicle design solutions.

Industry guidance sources do not focus on vehicle design specifically; they are more inclusive, with strong emphasis on policies and procedures. This holistic approach is essential but makes it difficult to delve into a high level of detail for vehicle design strategies. For example, recommendations from the CDPH, CalSTA, and Cal/OSHA include: “Where possible, install Plexiglas or other appropriate barriers in transit and rail vehicles to minimize exposure between operators and passengers” ( 24 , p. 10). This leaves a lot to be determined by the operator, for example barrier size, material, location, and whether to install barriers between passengers, all of which vary considerably in practice.

Reports of best practices ( 25 , 26 ) are useful supplements to general industry guidance. For example, the Federal Transit Administration released a resource to share practices that have been implemented by bus and rail operators worldwide during the pandemic, providing links to transit agency websites and news articles where operators can see specific examples of a wide range of strategies ( 27 ). Though immensely useful, best practices do not include emerging innovative design concepts that have not yet been implemented. APTA notes, “Due to the magnitude of the COVID-19 pandemic, it is likely that transit agencies will want viral outbreak considerations to be included in future vehicle and facility designs and modifications” ( 28 , p. 15). Broadening the scope to future vehicles creates room for more radical design innovations that should be considered alongside retrofit strategies to understand the universe of relevant design strategies to cope with this pandemic as well as other existing and future infectious diseases.

The global COVID-19 pandemic has given rise to a plethora of ideas for modifying and redesigning public transportation and shared mobility vehicles to protect workers and riders from contracting the disease while traveling. Public transit and shared mobility operators and their regulatory bodies are faced with the challenge of deciding which of the myriad proposed strategies to adopt, with a dearth of specific research on their effectiveness. These decisions are further complicated by the need to balance them with other types of strategies (e.g., station design strategies, policies, new business models, website and app features, and new cleaning protocols) to make up a portfolio of complementary solutions to recover ridership and keep workers and riders safe. All of this must be accomplished within financial limitations resulting from drastic revenue losses.

Research Aims

This research develops a typology of vehicle design strategies to mitigate COVID-19 risks. The typology is inclusive of strategies applicable to one or more shared or pooled travel modes, and inclusive of design concepts involving retrofitting existing vehicles as well as ideas for more radical redesigns for original manufacturing of vehicles. The purpose of the typology is to organize the myriad potential design strategies in a way that enables researchers, policymakers, and public transit and mobility service operators to more systematically and efficiently evaluate them and, in the latter case, incorporate them in a broader portfolio of solutions to protect their workers and riders.

A secondary contribution of this research is an exploration of expert perspectives on the identified vehicle design strategies, with an emphasis on prioritizing functional categories of strategies (e.g., those that mitigate the risk of fomite transmission [through contact with surfaces] versus airborne transmission [through droplets and aerosols]) and identifying knowledge gaps. Precedent for this methodology can be found in Zhang et al. ( 29 ), in which Zhang et al. surveyed transport and related experts about their opinions on the impact of COVID-19 on the transport sector and corresponding policy measures. Zhang et al. also reviewed 40 other studies that gathered expert opinions in the context of transport policy, noting that “expert opinions are useful to understand difficult issues, where limited knowledge is available,” which applies this new and complex issue of vehicle design strategies to help prevent COVID-19 transmission.

This paper does not make specific recommendations. Expert opinions are not a substitute for objective evaluations of the effectiveness of vehicle design strategies in mitigating the risk of COVID-19 transmission, which are crucial and are beginning to emerge ( 30 – 35 ). This research was part of a broader study that also included interviews with a small, convenience sample of public transit and shared mobility users, the details of which can be found in the report produced as a deliverable for the funded project ( 36 ). The following sections detail our data collection and analysis methods for the findings presented here.

Methodology

We analyzed vehicle design strategies in iterative stages based on the qualitative content analysis method described in Hesse-Biber and Leavy’s ( 37 ) 2010 publication. We conducted systematic internet searches, gathering information from news articles, public transit websites, industry reports, and academic literature. The literature search was conducted on Google and Google Scholar using combinations of three types of search terms: 1) shared and pooled vehicle and service types (e.g., “bus,”“subway”); 2) “COVID-19” or “coronavirus”; and 3) terms related to design strategies. The latter began with general terms, such as “vehicle design,” and expanded to sets of more specific terms as design strategies were identified (e.g., “seating layout,”“ventilation”). Researchers relied heavily on popular media and industry reports since there is yet limited scholarly research on this new topic.

Searches targeted all major shared and pooled modes: airplanes, buses, car-sharing, carpooling/vanpooling, ferries, ride-hailing, shared bikes and scooters, shuttles, taxis, trains and subways, trollies, and water taxis. The focus was on physical aspects of vehicles (i.e., space configuration, onboard features, equipment, and equipment settings/operations), including original equipment and retrofit strategies, and temporary and permanent features either affixed to the vehicle or located onboard the vehicle during operation. Design strategies range from innovations still in conceptual stages, to pre-existing technologies or settings that were already common or available (e.g., High Efficiency Particulate Air [HEPA] filters, windows open) but are now being highlighted as protective measures against the spread of COVID-19 in shared or pooled vehicles. Excluded from the analysis were physical design strategies found at transit stations but not in vehicles, policies (although some vehicle design solutions complement policies), business strategies, transit app and website features, cleaning services (using products or equipment brought onto the vehicle outside of its operation), and rider and worker personal equipment brought onto the vehicle that was not supplied by the service operator (including masks, gloves, and no-touch tools to open doors).

Vehicle design strategies were recorded with descriptions, images, and references to the media, industry, and/or academic articles that discuss them. Each strategy was also coded with the applicable vehicle/service type(s) mentioned in the sources (only those mentioned in sources and not theoretical considerations about potential applicability), and whether the strategy had been implemented yet according to the sources (as opposed to a conceptual design or under development). Data collected on each individual strategy are available online in the Supplemental Material, which also provides links to sources for each identified strategy.

Vehicle design strategies were organized into a COVID-19 risk-mitigating vehicle design (CRVD) typology. We attempted to create exhaustive and mutually exclusive categories based on considerations of structural (e.g., physical form and location in the vehicle) and functional characteristics (e.g., outcomes or services provided). We created graphics to illustrate all identified strategies (some graphics convey multiple specific, related strategies). The classification scheme and illustrations serve to summarize and distill the variety and complexity of observed vehicle design strategies into main ideas and themes that can be more easily communicated to policymakers, transit service operators, and vehicle manufacturers.

Interviewing Experts in Related Fields

After we inventoried the design strategies, we conducted in-depth, semi-structured interviews with experts in fields related to COVID-19 and its transmission. The aim was to explore their perceptions, at a high level, of relative effectiveness between and within categories of design strategies based on their expertise in related fields (e.g., infectious disease and airborne pathogen exposure). Equally important, we aimed to discover what these experts did not know about potential effectiveness of the design strategies, as implications for future research priorities could flow from these knowledge gaps. All interviewees were affiliated with University of California, Davis (UC Davis).

We interviewed 23 UC Davis faculty, research, and administrative staff with expertise in fields related to COVID-19, its transmission, or both, including: five physicians and three administrative staff at UC Davis Department of Internal Medicine Division of Infectious Diseases and Infection Prevention; one respiratory toxicologist; three engineers whose foci include airborne pathogen exposures; six engineers who specialize in HVAC, related controls systems, or both; three epidemiologists, two of which had expertise in air pollutant exposures and one in environmental justice; and two social scientists with public health and communications backgrounds, both conducting research on public perceptions and response to COVID-19. Interviewees included persons in senior leadership roles at relevant offices and divisions within UC Davis Health and several Organized Research Units (i.e., research centers).

Interviews were conducted via Zoom and most were one-hour long. The CRVD illustrations (see Supplemental Material) and mitigation mechanisms (Figure 1) were organized into a Google Slides presentation to screen share with participants during the interviews. We asked interviewees their perspectives on the potential effectiveness of the design strategies (with regard to the general categories and the specific strategies within each category) and other potential benefits and drawbacks of the strategies unrelated to COVID-19. We asked experts to discuss the mitigation mechanisms and give a sense of how they would prioritize them, for example, top three, or most and least important. We did not cover all CRVD categories with all experts. For example, we focused on the categories related to airborne transmission when we spoke with air quality and space-conditioning experts. For all others, we typically covered all categories unless we needed to shorten the interview to less than an hour based on interviewee time constraints.

Figure 1.

Risk-mitigation mechanisms.

CRVD Typology

Identified strategies were organized into the CRVD typology consisting of 12 main categories (Table 1). An Appendix provides a series of graphics to illustrate each category of CRVD strategies. Strategies were further analyzed to articulate 12 possible mechanisms by which they may help diminish the risk of contracting COVID-19 while riding in shared and pooled vehicles (e.g., increasing distance between passengers; Figure 1). The following sections overview each category, including examples and hypotheses about the mechanisms by which they may reduce risk, noting again that this is an inventory of design concepts and not a set of recommendations.

Table 1.

CRVD Typology Main Categories and Definitions

Vehicle design strategy type	Definition
Seating configuration	Seating layout and other features that specify where riders sit or stand during transit, including location and orientation
Pathways	Features that specify how riders move about the cabin, for example boarding and deboarding
Barriers	Partitions of various sizes, configurations, and materials between passengers or passengers and operators
Ventilation and air circulation	Equipment or setting that improves ventilation (brings outside air into the vehicle, moves inside air out) or air circulation (movement of air within the cabin)
Air filtration and cleaning	Equipment that cleans the indoor air that is being recycled or introduced from outdoors, by trapping and/or killing airborne particles
Onboard surface sanitization	Equipment and settings that implement surface cleaning processes
Hygienic materials	Materials that contribute to vehicle hygiene
Hygienic construction	Construction techniques that contribute to vehicle hygiene
Touchless technology	Adaptations or new mechanisms, including automated technologies, that limit driver/rider physical contact and interaction with the vehicle
PPE and supply provisioning	Onboard PPE and supply dispensers
Communication and monitoring	Features that collect and/or display information to passengers and/or service providers
Multimodal support	Features that facilitate multimodal trips to limit time or additional rides in shared/pooled modes

Note: CRVD = COVID-19 risk-mitigating vehicle design; PPE = personal protective equipment.

Seating Configuration

Seating Configuration includes three general strategies: eliminating seats, spreading them out, or changing their orientation. Eliminating seats can increase distance between passengers and reduce overall vehicle occupancy. Seats and standing passenger spots may be designated as approved or off-limits using signs or stickers, without any real structural changes. Seats may be temporarily removed or repurposed through flexible structural changes (e.g., removable seats). There are also proposals for permanent new layouts that include a reduced number of seats.

Unlike eliminating seats, spreading out or changing orientation do not necessarily require reducing occupancy. Three methods of spreading out seats (or passengers) were identified: designating a seating area (e.g., train car) for vulnerable populations to keep them away from others; spreading seats out into spaces not typically used for seating; and staggering/off-setting seats within rows. These all involve increased physical distancing.

Reorienting seats, including flipping middle seats to face the opposite direction to the adjacent row seats, and removing tables in train cars to flip alternating rows so all face the same direction, might increase distance slightly, create some degree of physical separation (more so if used in conjunction with a Barriers strategy), or perhaps redirect passengers’ respiratory particle trajectories away from each other’s faces (mechanism: Divergent Orientation). A differently motivated seating reorientation strategy is to use longitudinal bench seating instead of transverse arrangements for easier cleaning (mechanism: Surface Hygiene).

Pathways

In addition to specifying or changing where passengers sit (or stand) during the ride, vehicle design strategies can specify or change how passengers move about the cabin, for example, while (de)boarding. These strategies also leverage distancing, orientation, or both. Some bus services are roping off the front of the bus and requiring passengers to use the rear door only, limiting their proximity to the driver. On the other hand, some vehicles are designating separate doors for entry versus exit. Other Pathways strategies include an open gangway layout so passengers can more easily move from one train car to another (perhaps less crowded), and relocating ticketing machines in buses, where passengers linger before finding a seat, to be further away from the driver.

Barriers

Most Barriers strategies are after-market partitions added to vehicles on or near seats. They vary in terms of location (e.g., between individual passenger seats, above passenger or driver seat rows, around a driver), size and coverage (e.g., at the head area only, floor-to-ceiling), and material (e.g., soft plastic, plexiglass). These characteristics were used to come up with a naming convention for the distinct types of Barriers identified: e.g., Passenger Row Curtain (located above a rows of passenger seats and made of soft plastic), Driver Face Shield (located at the face area of the driver seat and made of hard plastic). Another set of strategies involves the use of seats as barriers, which includes a design concept where seatbacks are tall and there are no gaps between headrests, as well as podlike seats with wrap-around sides that separate adjacent passengers. Barriers are intended to block people from others' expelled respiratory particles.

Ventilation and Air Circulation

This category consists of equipment or system settings that improve ventilation (exchange with outdoor air) and circulation (movement of air within a space). Strategies to improve ventilation include opening windows, doors, and roof hatches, even when heating or air-conditioning; running the fan continuously and at maximum speed, using fresh (outdoor) rather than recycled air HVAC settings; and adding/using vents and fans, including exhaust fans. There are also recommendations to control airflow within the vehicle strategically, including directional airflow from clean to less clean areas, configuring open windows to create cross-ventilation flow, and indirect airflow from floor vents upward instead of horizontal flows directly into riders’ faces.

Air Filtration and Cleaning

For air filtration, advanced mechanical filters are widely recommended, including HEPA filters and filters with a high Minimum Efficiency Reporting Value (MERV) rating (e.g., MERV 13). These can be installed in the HVAC system or used in an after-market device added to the vehicle (e.g., a cupholder-sized air purifier with a HEPA filter). Other air-cleaning technologies include ultraviolet (UV) light (specifically UVC, also known as germicidal UV) and ionization, which can be integrated into the HVAC system or provided by an after-market device. UVC is extremely damaging to human eyes and skin, so when used outside of the HVAC ducting it needs to be enclosed in a system located on the ceiling that treats the air at the top of the vehicle without exposing riders.

Onboard Surface Sanitization

Onboard Surface Sanitization involves equipment installed in the vehicle that controls cleaning processes, as opposed to cleaning services conducted manually or with equipment brought onto the vehicle during servicing. Four onboard cleaning methods were identified: light, heat, chemical, and air. UV and LED lamps have been installed in vehicles, intended to kill the virus on surfaces between transit services. Foggers have been installed to spray chemical disinfectants between uses. Adapting an HVAC system to heat the vehicle cabin to a very high temperature is another strategy intended to kill the virus on vehicle surfaces between uses. Positive air pressurization of a bus cabin has been tested as a strategy to prevent the virus from settling on surfaces during vehicle operation.

Hygienic Materials

Recommendations can be found for easy-to-clean surface materials, including to avoid or remove cloth (e.g., rugs and fabrics on seats) and other porous materials, and use protective coatings, including clear-coat floor finishes, to facilitate cleaning. Relatedly, durable materials to hold up to harsh and frequent cleaning have been recommended, as well as disposable seat and floor coverings. Biocidal materials (e.g., copper) and coverings (films and spray-on shields), that is, those with internal properties that might eradicate the virus, have also been suggested. One source suggested replacing plastic surfaces with cardboard.

Hygienic Construction

Two general categories of construction techniques have been considered. The first is minimizing seams and joints, including non-ribbed flooring, sealing floor seams, and upholstering seats so that the bottom and back connection is seamless to avoid a gap where dirt and germs can accumulate. The second is accommodating cleaning, including detachable food trays, cantilever seating (easier to clean under), planning space for cleaning equipment to come onboard, floor-mounted piping to hook up a hose to clean, and holes in the bottom of seats (as drains for wet cleaning).

Touchless Technology

Touchless Technology strategies are divided into three sub-categories: automation strategies that eliminate the need for physical contact with the vehicle, reduce vehicle occupancy, or both; low-touch mechanical strategies that minimize physical contact; and personal props that involve replacing shared surfaces with accommodations for personal items. Examples of automation include fully automated (i.e., driverless) vehicles, touchless card and ticket scanners, automated doors, and automatic hand-sanitizer dispensers (rather than pumped). Examples of low-touch mechanisms include a pedal door opener and pedal tray table control. Personal props include personal handholds distributed to passengers to attach to vehicle grab rails to avoid contact with the shared surface, and mounts in planes for personal mobile electronic devices and bags, replacing shared touchscreen tablets and seat-back literature pockets, respectively.

Personal Protective Equipment and Supply Provisioning

Service operators have added dispensers to their vehicles to supply users with hand sanitizer, disinfecting wipes, and masks. In addition to onboard dispensers, many operators are supplying personal protective equipment (PPE) to their workers. Other related strategies include locating trash receptacles in vehicles (e.g., to dispose of wipes), and making sure onboard restrooms are adequately stocked with soap and dryers or paper towels. These strategies are indirectly related to the risk-mitigation mechanisms listed in Figure 1, since the mediating role of occupant behavior is more significant; however they do have implications for Surface Hygiene, and Physical Separation in the case of masks.

Communication and Monitoring

Communication and Monitoring strategies are divided into three sub-categories: education and prompts; environmental feedback; and symptom detection and contact-tracing. Education and prompts remind passengers to comply with the federal mask mandate and other operator policies and recommendations (e.g., to practice good hygiene), as well as communicate CDC guidance and general COVID-19 information. Environmental feedback involves devices that monitor and display dynamic information about environmental conditions in the vehicle related to the risk of virus transmission, such as occupant density, surface cleanliness, and carbon dioxide (CO₂) levels. Symptom detection and contact-tracing strategies collect data to monitor passengers. Examples include a QR code in pooled vehicles that passengers scan to enable contact-tracing in the case of exposure to another passenger with COVID-19, and cameras with facial recognition to detect mask-wearing or thermal imaging to detect whether a passenger has a fever.

Multimodal Support

A unique feature of a train-car design concept is the flexible repurposing of some seating as convenient bike storage for passengers. This Multimodal Support strategy could facilitate or encourage the use of bikes in conjunction with pooled and shared travel modes to shorten the duration of trip legs made in pooled vehicles where exposure risk is higher. Although no mentions of added bike racks for this purpose were found for other types of shared and pooled vehicles, this strategy is also relevant for cars (taxi, ride-hailing, and vanpooling) and buses.

Expert Opinions

Expert interviewees tended to consider the potential effectiveness of CRVD strategies and mitigation mechanisms in the context of the two general modes of COVID-19 transmission: fomite transmission (through touching contaminated surfaces) and airborne (droplet and aerosol) transmission (respiratory particles landing in mucous membranes of the eyes, nose or mouth, or being inhaled). Most CRVD categories fall into one of these meta-categories, having implications mainly for one transmission mode or the other. Thus, we used these categories to organize the analysis below, followed by a third category of “passenger-mediated strategies” that rely on user behavior to be effective (via mitigation of fomite or airborne transmission).

Airborne Transmission

Five CRVD categories are primarily related to airborne transmission: Seating Configuration, Pathways, Barriers, Ventilation and Air Circulation, and Air Filtration and Cleaning. These strategies might reduce the likelihood of larger, heavier respiratory droplets from one rider landing on another rider’s mouth, nose, or eyes, and/or reduce the likelihood of riders inhaling smaller particles. Experts explained that droplet and finer aerosol transmission modes are difficult to tease apart but there was a consensus that CRVD strategies that address aerosol transmission risk in particular have the greatest potential for preventing transmission. Chief among these are space-conditioning strategies (i.e., Ventilation and Air Circulation; Air Filtration and Cleaning).

Experts overwhelmingly ranked Increased Air Exchange as the top priority mitigation mechanism, with some grouping it with closely related mechanisms Air Cleaning and Strategic Airflow. One summarized, “Dilution is always the solution to pollution.” Others warned, “If you don’t have high airflow, you’re missing the boat” and, “If you don’t have sufficient ventilation overall there could be a large background [of respiratory particles] building up for indirect exposure.” They noted that Separate Air Spaces would be highly effective but there are limited applications for that in pooled travel; exceptions include operator compartments that are sealed off from the rest of the cabin and full barriers between the front and back seat in taxis and ride-hailing vehicles, but these would have to be used in conjunction with space-conditioning strategies that keep the air in these spaces separate while not cutting anyone off from fresh air ventilation and heating and cooling.

Social science experts pointed out that some of these strategies to reduce risk may be invisible, unfamiliar, or both, to riders. They suggested including an educational component in such cases, to communicate how the strategies work. When people know about these strategies it may give them “peace of mind”/make them “feel safer.”

Ventilation and Air Circulation

The general sentiment was the more fresh/outdoor air the better (e.g., open-air shuttle being the epitome of good ventilation), with some caveats. Space-conditioning experts mentioned that maximizing ventilation could make the ride less comfortable or interfere with aerodynamics depending on vehicle speed and weather, and increase energy consumption if also heating and cooling. Bringing in outdoor pollution, for example, from the roads or underground subway systems, was also a concern among some health experts; some advised that incoming outdoor air should be filtered. One ventilation strategy that was met with some skepticism was the headrest air vent idea. One expert suggested that the headrest might be better placement for exhaust to suck the air out before it goes from the front to the back seat; another wondered about the ducting situation to hook the vents up; and a third noted that you might need to resize the HVAC fan to increase power, otherwise all vents would have reduced flow. A few also wondered whether the bus ventilation strategy that directed air from the front to the rear, up and out the roof might leave the passenger seating area stagnant.

One HVAC engineer noted that it is important to have airflow at people’s heads and they liked how planes have the vents streaming right at passengers’ faces; another relatedly remarked that the ideal situation when recirculating and filtering air would be to inhale right from the vent. They thought the indirect airflow strategy (vertical airflow in front of passengers rather than horizontal airflow at head level) made sense. Concerning the best direction of vertical airflow, some suggested upward movement would complement the natural rising of human breath that is warmer than the indoor air and could be effectively combined with exhaust at the top of the vehicle. One the other hand, one expert pointed out that, “Airflow in vehicles is much more driven by the HVAC than buoyancy (hot air rising). If hot air rising is a factor, then your air exchange is too low.” And another expert thought it might make more sense for road vehicles to draw cleaner air from above rather than below the vehicle.

Space-conditioning experts explained that airflow is extremely complex, with many determining factors, and thus it is difficult to control. Some also considered the difficulty of objectively evaluating these strategies, noting that computational fluid dynamics studies would need to be done for each specific vehicle in a variety of contexts—and even then, it would be hard to predict reliably. The unifying ideas across expert opinions involved plug flow (a type of laminar airflow with little to no turbulence to avoid mixing contaminated and fresh air) moving in one direction from clean (unoccupied) to less clean (occupied) areas where exhaust outlets are located, such that particles expelled from a potentially infected person will not be entrained in a flow that goes into someone else’s face and will be quickly moved out of the vehicle.

Air Filtration and Cleaning

Space-conditioning and air-quality experts explained that air filtration and cleaning strategies (specifically mechanical filtration and UV germicidal irradiation in HVAC system ducting or upper-room systems) can function similarly to enhanced ventilation in respect of increasing air exchange, with some recommending just that code-required ventilation be ensured and then supplemented with enhanced air cleaning. Experts familiar with ionization disapproved of it, referring to research showing a lack of effectiveness and concerns about generating other air pollutants (i.e., ozone, albeit some thought perhaps in negligible amounts); one worried that incorporating ionization technologies might contribute to a false sense of security. Space-conditioning experts also noted that incorporating any new cleaning technology into the HVAC system introduces more maintenance requirements and room for error. Introducing after-market/portable air purifiers seemed a good idea to most, with caveats including that they are intended for use in closed spaces, need to be sized appropriately for that space, and products range in quality: “It’s important to be cautious … buying purifiers …. You want one with an industrial or FDA claim (medical grade).”

Barriers

Experts mentioned the distinction between droplets and aerosols when considering the potential effectiveness of barriers, noting that they are generally more useful to protect from the former than the latter. Some experts described how barriers are more useful if people are going to be interacting/talking (i.e., expelling droplets) and so barriers to block workers from passengers who would be talking to them were a good idea. Some observed that hard barriers are preferable to soft materials for ease of cleaning and reliability (easier to keep in place).

Experts saw barriers as less important if masks are worn (properly) since masks serve a similar function but more effectively since they are closer to the potential exposure site. However, many also suggested that barriers might have a “psychological impact,” making passengers feel safer. Some experts indicated the importance of interactions between barriers and space conditioning, including a concern that some barriers (particularly those perpendicular to airflow or extending over passengers’ heads) may create stagnant air pockets that could increase aerosol transmission of COVID-19.

Seating Configuration and Pathways

Experts felt that reducing vehicle occupant density (i.e., the number of passengers) would be more effective than strategies that only increase physical distancing or those that reorient the direction passengers face while seated or during vehicle boarding and deboarding. They explained the importance of reduced occupant density in relation to aerosol exposure risk, noting that ventilation requirements are based on occupant density. One summarized, “You can’t ventilate your way out of crowding.” Another explained, “It comes down to occupancy density… The more people you have, the density of aerosols increases… Imagine every person has a bottle of perfume with a cap off… When you walk in the bus, you’re going to be like, ‘Whoa!’ It doesn’t matter how far apart people are.” Some experts also considered how seating arrangements might interact with airflow. For example, one considered, “Someone is always going to be in a bad seat.” This related to two ideas: one is that airflows might move across passengers (e.g., from the front to the back of the vehicle) before exhausting so the person in the last seat in line gets the most exposure, and the other is the idea that aisle seats may have greater exposure from people walking up and down the aisle.

For Reduced Occupancy, most interviewees thought that temporary or flexible solutions were more practical than permanent designs with fewer seats since decreasing ridership would be difficult for service providers and prices might increase for riders; a few also mentioned the negative environmental impact of increasing the number of vehicle/trips to provide the same level of service. A few observed that temporarily removing or repurposing seats would be better than relying on signage and ropes designating seats as off-limits that passengers might ignore. Some interviewees pointed out that certain Seating Configuration strategies would not allow groups to sit together. A few had concerns that the flipped middle seat might increase risk since people would be facing each other (the middle seat would now face the window and aisle seats of the row behind) which might encourage talking, and it could cause motion sickness.

Fomite Transmission

Four CRVD categories are primarily related to fomite transmission: Onboard Surface Sanitization, Hygienic Materials, Hygienic Construction, and Touchless Technology. Experts explained that these strategies would have limited value to mitigate the risk of COVID-19 transmission because fomites are not the main mode of spread. One expert referred to COVID-19 fomite transmission as a “red herring…It isn’t that it helps 0%, but why waste resources?” Another similarly observed, “It’s not a no-yield, but not a priority. I wouldn’t direct my resources [to these strategies] but would not ignore [them] if easy to implement.” When asked to (roughly) rank the risk-mitigation mechanisms, virtually all experts placed those related to fomite transmission (Avoided Surface Contact and Surface Hygiene) as least important.

Although interviewees felt these strategies should not be high priority concerning resource allocation for short-term solutions to the current pandemic, medical experts pointed out that other diseases (including perhaps the next pandemic) do spread easily through fomites, one noting that “shared surfaces are always going to be an issue.” They felt that these strategies would be relatively more important for high-turnover vehicles where there is little opportunity for cleaning in between services. Experts also thought shared and pooled mobility users might appreciate these strategies since they can be highly visible and easily understood. One medical expert considered patients’ reactions to robots they have in the hospital using UV technology to clean surfaces: “People see UV robots in the hospital, and you explain what it’s doing, they immediately feel a lot safer.” However, one expert considered how explaining fomite-related prevention measures might do more harm than good in the context of COVID-19, questioning, “Do you actually end up hurting people because they pay more attention to those? [They think,] ‘I’m doing that, I’m fine.’”

Touchless Technology

Interviewees generally found automation strategies useful beyond the context of COVID-19, for hygiene and accessibility reasons. They had some criticisms about the low-touch strategies, remarking that foot pedals may not be accessible to everyone or may get in the way, and hooks and stands for personal devices would still be shared surfaces. One health expert was passionate in her disgust for seat-back pockets in airplanes because of hygiene concerns, so the idea of their removal was happily met, but again not because of particular relevance to preventing COVID-19 spread.

Hygienic Materials and Construction

Medical experts confirmed the value of durable materials to hold up to harsh cleaning based on experiences with hospital equipment, although some mentioned possible sacrifices to comfort during long trips on hard seats. Medical experts were also familiar with biocidal materials and validated their potential but also raised concerns such as lack of evidence of effectiveness outside laboratory contexts, copper and cardboard being difficult to clean, and surface contamination interfering with light-activated self-cleaning materials. Some mentioned the waste associated with disposable seat and floor coverings.

Onboard Surface Sanitization

Health experts were generally familiar with these strategies, except for positive pressure. One (physician) noted that routine cleaning has been shown effective against even the more virulent strains of COVID-19. Another (epidemiologist) considered that the use of heavy chemical cleaning may be warranted under conditions of high community spread of a virus with significant fomite transmission but should be discontinued after the threat subsides. Concerns were also raised about potential harm to vehicle surfaces (from UV, heat, and chemicals).

Passenger-Mediated Strategies

Three CRVD categories were less strictly tied to either fomite transmission or droplet/aerosol transmission: PPE and Supply Provisioning, Communication and Monitoring, and Multimodal Support. What these three have in common is a greater degree of dependency on rider behavior. Their effect on either fomite or droplet/aerosol transmission risk is indirect, mediated by actions taken by passengers (e.g., using provided PPE, following posted guidance, or bringing their bike on the train and getting off a couple of stops early). Riders can have influence on other strategies (e.g., rolling down a window for ventilation or sitting in a seat marked as off-limits), but these three categories are more exclusively under the riders’ control.

PPE and Supply Provisioning

Interviewees thought PPE and Supply Provisioning was a good idea, although one medical expert pointed out that these strategies are not considered PPE in the medical community (e.g., their definition would include things like N-95 masks after a medical fit test). Medical experts who interact regularly with the public described how hand-sanitizer use has become routinized for many people and public dispensers are widely used, to the point where some people have developed preferences (and distastes) for certain types of sanitizers. However, one expert noted that hand sanitizers can give off volatile organic compounds (VOCs), so being in a vehicle where many people walk in and use sanitizer could be harmful, particularly for vulnerable groups (e.g., individuals with asthma). There was also some concern about disinfecting wipes and masks generating waste (this was mentioned for floor/seat coverings as well). One expert observed that these strategies are not only helpful in preventing the small risk of surface transmission, but also help to create a “prevention and safety culture.” Another suggested that pairing these strategies with “ads or announcement over the PA systems” could be effective in encouraging desired behaviors, which relates to the next section.

Communication and Monitoring

Most interviewees thought prompts to abide by policies and recommended practices were useful reminders, though not high-impact strategies. One explained that these strategies are important when guidance is inconsistent across settings and directives are changing. Quite a few also pointed to the utility of these strategies as an aid to help employees enforce the rules (giving them something to point to when they have to ask someone to comply) and to help passengers apply social pressure to someone not conforming.

For messaging formats, some interviewees mentioned that basic signage can be overlooked, whereas audio announcements can be more salient but sometimes annoying. Experts in communication and social science stressed the need to use multiple languages and clear graphics as an aid for language and literacy barriers. They also recommended these strategies for educational messages about invisible or unfamiliar strategies, such as air filtration, with one noting that varying levels of literacy is particularly challenging when trying to explain complicated technologies (e.g., HVAC) and scientific evidence (e.g., general information about disease transmission).

Symptom and mask compliance detection and contact-tracing strategies were largely met with skepticism about reliability and feasibility of enforcement. Medical experts had experiences with thermal imaging and related technologies and were unimpressed with their accuracy and validity. Several suggested these occupant-monitoring strategies would be highly unpopular among many Americans because of perceptions of privacy infringement. Some pointed out that certain (and more vulnerable) populations (e.g., older adults) would not be adept at using QR codes.

Experts thought environmental feedback strategies could be well-received since these give riders more agency, for example, “so they can make their own decisions and assessments.” Experts discussed the need to provide ample education with these strategies, particularly CO₂ feedback, and opportunities to respond, for example, move to a less crowded train-car or find a cleaner seat. In the absence of control, interviewees said environmental feedback “is going to make people more panicked.”

Multimodal Support

Most interviewees thought the one specific strategy in this category (bike racks—adding them, improving accessibility, or both) could help reduce risk if people reduced the amount of time spent in the pooled vehicle, with some referencing the CDC guidance on time limits for reducing indoor exposure risk (i.e., 15 min). However, many had doubts about how much it would encourage riders to shorten their trips. Most thought it was a good idea beyond the context of the current pandemic, to facilitate biking, with some health experts referencing the importance of exercise.

Discussion

This research distilled the wide variety of vehicle design solutions aimed at mitigating the risk of COVID-19 transmission in public transportation and shared mobility into the CRVD typology. The CRVD typology outlines a research agenda. Environmental exposure scientists and other experts can study and compare the effectiveness of strategies across and within the CRVD and mitigation mechanism categories and identify gaps in existing relevant literature. Social scientists and travel behavior researchers can assess the influence of CRVD strategies on worker and rider attitudes, intentions, and behaviors. CRVD strategies with the greatest potential impact in mitigating spread are top priority, but it is also important to address user perceptions and consider strategies that help workers and riders feel safe.

In-depth, semi-structured interviews revealed a consensus among experts in fields related to COVID-19 and its transmission that CRVD strategies that mitigate aerosol transmission have the most potential to reduce risk, which aligns with the current state of knowledge about COVID-19 ( 38 , 39 ). This suggests a general prioritization for objective evaluations of CRVD strategies; specifically, more studies are needed to understand how space conditioning (Ventilation and Air Circulation, Air Filtration and Cleaning), Barriers, and Seating Configuration in shared and pooled vehicles affect COVID-19 aerosol transmission. These studies have started to emerge ( 30 – 35 ) and are crucial to provide specific guidance for public transport and mobility services operators. Since Air Filtration and Cleaning strategies are less familiar and generally less observable to users, these should be coupled with educational strategies to make users aware that they are in operation and explain how they work (this applies also to Ventilation and Air Circulation measures that are not readily perceived by users, e.g., when mechanical ventilation is supplying fresh air).

Experts suggested that resources devoted to CRVD strategies be concentrated on those with the greatest potential for risk mitigation (i.e., those geared toward reducing aerosol transmission). They recognized that some strategies could be quite expensive for operators to implement across their fleets, requiring new equipment or reduced service capacity (e.g., removal of seats), and if large investments are made, they should “focus on what matters.” Another consideration is the permanence of CRVD strategies; it is difficult to weigh the value of more radical and permanent solutions, and to know when to revert temporary changes back to pre-pandemic arrangements, when there is much yet unknown and changing about the nature of COVID-19.

Experts felt that strategies geared toward reducing fomites, though generally desirable, would do little to mitigate COVID-19 spread, and some feared that conveying them as COVID-19 mitigation measures might create a “false sense of security” whereby passengers and workers might ignore more effective measures. This is a question for future study, though research on similar types of risk compensation (e.g., effects of mask-wearing on hand hygiene) suggests this may not be a serious concern ( 40 ). Educational strategies within and beyond the realm of transit and mobility services should (continue to) communicate, in layperson’s terms, the latest scientific findings about how COVID-19 is transmitted (and prevented).

Transportation service providers, researchers, and policymakers should consider relationships among CRVD strategies when identifying solutions. Experts highlighted several kinds of relationships among CRVD strategies and mitigation mechanisms more generally: surrogacy, combination, and interaction. For example, some observed that a similar but less effective strategy could serve as a surrogate solution when the more effective strategy is not feasible, for example, if you can’t increase Ventilation, add Air Filtration, or if you can’t increase distance between passengers (Seating Configuration), add Barriers. Many pointed out that combinations of strategies are optimal, for example, Ventilation plus Air Cleaning; “Good solutions are a combination of materials and cleaning”; “Ads or announcements over the PA systems [in combination with PPE and Supply Provisioning] could encourage good behaviors.” An example of interaction is determining whether a higher-efficiency HVAC filter (trapping more particles) would sacrifice (i.e., reduce) airflow speed to the point that a lower-efficiency filter would provide a more optimal balance of air exchange and air cleaning. Interviewees also discussed some dependencies between policies and design strategies, for example, most CRVD strategies would become more important if there were no mask mandate on pooled and shared modes of transportation.

Transportation service providers, researchers, and policymakers should also consider the potential for unintended consequences of CRVD strategies, maintenance and quality control requirements, and the ability to enforce measures reliant on human behavior. Experts had concerns across CRVD categories about the possible introduction of additional hazards, such as outdoor air pollution from enhanced Ventilation; chemical and UV exposure (Onboard Surface Sanitization); ozone (Air Cleaning by ionization); more COVID fomites (Barriers if not cleaned); VOCs (PPE and Supply Provisioning of hand sanitizers); and accumulation of other organisms (use of cardboard in Hygienic Materials). They also had doubts about the feasibility of implementing some strategies that would require increased maintenance and quality control measures (e.g., Air Cleaning and Onboard Surface Sanitization), or or /or enforcement measures (e.g., Communication and Monitoring, and Seating Configuration: eliminating seats through signage). One expert explained, “Best of course are interventions that don’t require repeated individual behaviors” (this could apply to both passengers and operators/employees); and another summarized, “You need to balance what’s best with what’s accomplishable.”

The CRVD typology can also be used by transportation providers and vehicle designers as a guide to generate more solutions, within and beyond vehicle design. For example, they can challenge themselves to identify more strategies of a given type such as Communication and Monitoring (e.g., onboard signs or audio messages explaining ventilation systems). They can also explore categories with few strategies (e.g., Multimodal Support) to see if more can be developed.

Limitations

The CRVD categories are intended to be exhaustive and mutually exclusive. However, additional categories may emerge as more design strategies are proposed and implemented. There are also a couple of areas of overlap among the categories, such as automated hand sanitizer dispensers and touchless trash cans, which fit in both PPE and Supply Provisioning and Touchless Technology categories.

Functional interactions and complementarities between different categories and between design and other types of strategies (e.g., policy) are not addressed in the CRVD typology, though some are discussed in the interview analysis. For example, signage (Communication and Monitoring) communicates policies to riders, which may be in service of other types of CRVD strategies, for example, designating a seat as off-limits (Seating Configuration). The typology could be expanded to include policies, station features, and service app/website features related to each category to support a more holistic approach to risk-mitigating pooled and shared mobility service design.

Some concepts design firms are pitching include a composite of strategies that cut across multiple CRVD categories. The typology breaks down these design concepts into discrete strategies, but operators may be procuring them as packages. Examples include a three-seat row design for planes with pod-like seats (Barriers) with the middle seat facing backward (Seating Configuration), and a flexible seating configuration for trains with seats that fold up beneath bike racks to temporarily repurpose the seat space (Seating Configuration) as micromobility storage (Multimodal Support).

It is important to note that the applicability and value of any given category or specific strategy may vary between vehicle and service types, since baseline conditions vary. In the interest of surveying the breadth of vehicle design strategies available across all pooled and shared travel modes, this research did not delve deeply into distinctions between vehicle and service types. Vehicle/service types for which each design strategy was applied (or proposed) in the identified sources were recorded but not necessarily exhaustive of all applicable vehicle/service types.

The expert opinions gathered and summarized here are not a replacement for objective evaluations of the effectiveness of vehicle design strategies in mitigating the risk of COVID-19 transmission. Expert interviewees almost universally qualified their expertise despite strong competencies in related fields. Space-conditioning experts all focused on building HVAC in their own research and had relatively less knowledge of vehicle HVAC systems. Medical and health experts stressed that our understanding of the pandemic is still evolving and there is a lack of evidence base for many of these strategies, particularly in relation to evaluation of clinical impacts of real-world applications as opposed to laboratory studies. That said, it is important to note that our interviewees stressed the dominant role of aerosol transmission of COVID-19 even though our interviews were conducted before the CDC and WHO were making such claims. Evidence has emerged since that backs up this perspective and lends credibility to our interviewees (e.g., Prather et al. [ 38 ] and Anderson et al. [ 39 ]).

Conclusion

This research developed a COVID-19 risk-mitigating vehicle design (CRVD) typology to analyze and summarize the wide variety of vehicle design strategies that have been implemented or suggested to reduce the risk of COVID-19 transmission among workers and passengers in shared and pooled vehicles, and explored expert perceptions of the identified CRVD strategies. The typology and expert opinions suggest a research agenda for objective evaluations of the effectiveness of these vehicle design strategies and their relationship to user preferences and travel behavior, both within and beyond the current context. Paired with emerging research on the efficacy of these strategies in vehicles and comparable contexts, as well as public health and industry guidance, the CRVD typology can serve as a reference to help public transportation and shared mobility operators assess the range of possible vehicle design solutions and determine which are suitable for their vehicles and services, in their continued response to the pandemic as well as for future planning. Ultimately, this research aims to help support a safe return to shared and pooled travel in the wake of the pandemic and contribute to a better—more equitable, sustainable, and enjoyable—mobility future.