As is well known [1], Infrastructure is defined as the physical system of an activity or region and often involves the production of public goods or related production processes. Critical infrastructures, CIs, refer to the ones that are essential for the functioning of a society and economy. Examples of critical infrastructures, CIs [2-5], include (see Fig. 2):

• Chemical Sector
• Commercial Facilities Sector
• Communications Sector
• Critical Manufacturing Sector
• Dams Sector
• Defence Industrial Base Sector
• Emergency Services Sector
• Energy Sector
• Financial Services Sector
• Food and Agriculture Sector
• Government Facilities Sector
• Healthcare and Public Health Sector
• Identifying Critical Infrastructure During COVID-19
• Information Technology Sector
• Nuclear Reactors, Materials, and Waste Sector
• Sector-Specific Agencies
• Transportation Systems Sector.
• Water and Wastewater Systems Sector.

In turn, the Transportation Systems Sector includes:

• Aviation
• Highway and Motor Carrier (critical road infrastructures, CRI)
• Maritime Transportation System
• Mass Transit and Passenger Rail sector
• Pipeline Systems
• Freight Rail
• Postal and Shipping.

ICT have a twofold nature: they are both CIs and critical information infrastructures, CIIs. They provide critical infrastructure protection, CIP, i.e., they are ICT networks that support, link, and enable other CIs.

To this end, it is noted that the European Programme for Critical Infrastructure Protection (EPCIP) has been laid out in EU Directives by the Commission (Commission of the European Communities).

Fig. (3). Examples of risks (3a, above) and relationships among safety, security, system, and infrastructure (3b, below). Symbols. AM = asset management. OBS = obsolescence. Sources [24–32].

Fig. (4). Risk-related terminology.

CI complexity, interconnectedness and interdependency amplify the concerns related to their protection and to this end, ICT infrastructures can be used to provide critical infrastructure protection as a set of “risk management and operational response activities aimed at improving the security and resiliency” of CIs [2, 6, 7]. From this standpoint, ICT (as systems of software, hardware, and services) help other CIs to be more resilient. Not only does ICT help protect other CIs, in terms of Critical information protection, but additionally ICT itself has to be protected in terms of cybersecurity and Critical information infrastructure protection, CIIP. In other words, CIIP refers to the activities aiming at protecting critical information infrastructures (e.g., related to transports, communication, finance, military, and energy) from cyber-attacks [8, 9]. Importantly, computer security (or cybersecurity) addresses both CIIP and CI and is a vital part of Homeland Security.

Fig. (5). Theoretical framework.

Fig. (6). Parameters influencing PA (eq. 2) and PST (eq. 3).

Transportation systems combine elements and their interactions that produce the demand for travel within a given area and the supply of transportation services to satisfy this demand [10]. Supply and demand relate to facilities (transportation infrastructures, e.g., roads, railways, and airports), users, vehicles, and environmental conditions. On the other hand, when it comes to supply, communication networks and transportation infrastructures are becoming more and more synergistic and it seems already difficult to think of several applications (e.g., infrastructure monitoring and autonomous vehicles) without a close relationship and synergy between communication networks and transportation infrastructures. For this reason, the term infrastructure (instead of transportation infrastructure) is sometimes used, where the infrastructure is meant to include communication and transportation “entities”.

The concept of risk [11, 12] related to a given “negative” scenario can be quantitatively determined by combining its probability of occurrence (typical or proper of the given scenario) and the magnitude/severity/extent of the related consequences (e.g., damage and/or loss that can be computed considering the expected number of fatalities in a given period of time or per unit of exposure time, the subject exposition, the expected damage, the vulnerability of the subject that is exposed to a given hazard, and the expected disutility). Other risk-related concepts are [12]: 1) Subjectivity (e.g., decision maker or user, person or society, etc.); 2) Time (e.g., risk related to today or yesterday, based on different boundary conditions); 3) Space (e.g., risk referred to here or there); 4) Uncertainty. Fig. (3a) illustrates nine examples of risks that refer to critical road infrastructures (CRI), including attacks (homeland security, left, 2a1: highway attack, 2a4: cyberattack; 2a7: bomb attack), natural disasters (centre, 2a2: earthquake; 2a5: flooding; 2a8: landslide), obsolescence- and decay-related failures (right, 2a3: pavement failure; 2a6: friction decay; 2a9: bridge failure). Fig. (3b) focuses on the relationship between asset management, (homeland) security, and safety, where asset management (including pavement management) mostly impacts safety (freedom from risk or harm as a result of unintentional acts) but could also affect security (freedom from risk or harm resulting from violence or other intentional acts). It seems relevant to highlight that land transport systems may be affected by different types of risks, pertaining to different components (Fig. 3b). The concept of road safety necessarily includes the consideration of users and infrastructures (probability of accident, P_A). In contrast, the concept of security may address, for example, the structural integrity of embankments, tunnels and bridges, P_ST, (infrastructure), or data integrity attacks in autonomous vehicle networks [13, 14].

Risk matrices [12] can be used to represent the relation between probabilities (columns of the matrix) and consequences (rows of the matrix), which grow from left to right, and from the bottom to the top, respectively.

Berdica [12] analysed road transportation system vulnerability from both a safety and a level of service sufficiency point of view, and partially reviewed the concept of reliability. Vulnerability can be defined [12, 15] as a susceptibility to accidents that reduce road network accessibility of the whole road system (i.e., not only of the physical network).

Outaly [16] demonstrated the efficacy of V2V and V2I communications towards reducing the risk of collision (as well as reducing CO₂ emissions). Similarly, Sepulcre and Gozalvez [17] presented V2V applications for traffic safety. These studies highlight that ICT is a tool to increase safety and reduce the probability of accidents. Liu and Fan [18] demonstrated the effectiveness of strain data in terms of impact on failure probability analyses of concrete bridge girders.

Frangopol [19] focused on the relationship between reliability and structural monitoring data. These studies point out that monitoring data provide information that reduces the probability of structural failure.

Finally, Troisi and Castaldo [20], outlined a theoretical framework for the risk management of road infrastructures from the perspective of organizational studies and engineering science. In more detail, they focused on the potential of satellite data obtained from differential interferometric synthetic aperture radar in a geographical information system platform.

In turn, accessibility [12] represents the opportunity (offered by the system mentioned above) to reach a given location and can be increased acting on the number of routes, or on the number of services reachable on a given route or within a certain time budget. Furthermore, accessibility [12] can be seen from both the demand and the supply sides, and the choice of point of view is important for the quantification of the consequences (e.g., delay). In particular, the first point of view (demand side) relates to the goodness of functioning of a given route, while the second one (supply side) deals with the existence of a functioning route that makes the journey worthwhile (i.e., a route/link/road that connects two locations in a given period of time, and at a given cost; a concept that is known as serviceability). Hence, the aforementioned consequences can be properly calculated if the second point of view is used. Negative scenarios/events [12] that affect the road transportation system’s vulnerability and serviceability include extremely adverse weather, the occurrence of physical failures (loss of structural integrity), traffic accidents, and intentional harm (e.g., wars or terroristic actions). Note that the vulnerability is connected to the consequences of negative scenarios/events, regardless of the probability of occurrence of that negative occurrence [12]. Another important concept is robustness [12]. It can be associated with the “ability of a system to withstand strain”, and is the opposite of vulnerability. At the same time, the concept of resilience [12, 21, 22] is often associated with the aforementioned systems and refers to the ability (in terms of speed and degree of recovery) of a system to “return to normal”, or “reach a new state of equilibrium”, or “restore the original serviceability level” after the occurrence of an event/perturbation. Usually, congestion is the main parameter used to address traffic issues [23]. With the growing increment of user mobility, the importance of reliability is substantially growing [12, 23]. Importantly, reliability can be seen as a sort of complement of vulnerability, i.e. “vulnerability in road transportation system is reliability, meaning adequate serviceability under the operating conditions encountered during a given period ” [12]. Reliability is linked to the concepts of (1) connectivity with the given destination (terminal reliability), (2) time (reliability of travel time), (3) probability to handle a given amount of traffic (capacity reliability), and (4) traffic flow (considering normal conditions, and recurrent and non-recurrent network congestions). It is possible to state that a road network is “reliable” [12] if the traffic service provided by the network, even in case of route interruption due to the occurrence of negative scenarios (e.g., accident, maintenance, natural disaster, etc.), is able to offer alternative routes to the users. There are several policy instruments to manage transportation networks' reliability [23], physical expansion of capacity, better management of capacity, pricing mechanisms to deliver a market for reliability, and use of information systems to mitigate unreliability. Despite this, reliability is rarely used as a parameter in cost-benefit assessments carried out during transport policymaking. Consequently, reliability targets and indicators should be selected with caution and taking into account both network operator and user perspectives [23].

Reliability, availability, maintainability, and safety (RAMS) analyses can be used to assess the performance of a system with respect to pre-defined requirements. RAMS analyses can be used for transportation systems [33, 34], although this approach is not comprehensive because of the fact it is not balanced from an economic point of view. Life Cycle Costing analysis can be used to fill this gap [34, 35]. The relationships among the diverse risk-related concepts analysed above are summarised in Fig. (4).

Table 1 and Fig. (5) illustrate the overall framework, i.e., the conceptual scheme to describe road risks taking into account both traditional and innovative methods and strategies.

By referring to critical road infrastructures, Traffic affects environmental and operational impacts, safety and efficiency of transportation systems, and can be analysed (under normal or congested conditions) using data from measurements and/or models. Importantly, traffic models should be always properly validated, and, to this end, micro-simulation models can be effectively used (especially during the calibration process [47]).

The risks of accidents can increase because of extrinsic factors (e.g., driving behaviour, driving dynamics, environment, vehicle speeds) and/or intrinsic factors (e.g., road geometry inconsistency) [37-39]. In particular, intrinsic parameters, such as horizontal radius, section length, and longitudinal slope, can be used to explain the entire safety process. For example, the horizontal radius affects road accidents' stochastic process, such as the section length. “Speed inconsistency” can be related to friction requirements (e.g., side friction coefficients). In such cases, road agencies have to make a consistent reduction of the speed limit [39]. To this purpose, a methodology for friction-based assessment of speed limits (based on braking distance and stability in curves in both wet and dry conditions) can be used [39]. Rasmussen [36] focused on the correlation between accident-related risk and the performance of decision-makers at different levels (i.e., from national to local government, policy and budgeting to company planning). He found that “accidents are created by the interaction of potential side effects of the performance of several decision makers during their normal work”.

Unsatisfactory levels of friction and texture or the decay of friction increase the risk of accidents. Friction is not the only property that can decay over time. In fact, especially for porous asphalts, bulk and functional characteristic undergo a reduction during the pavement’s expected life, affecting performance, safety, and budget [39-41]. In particular [42], with a dry surface, road safety mainly depends on micro-texture (which, in turn, is strongly affected by petrology) and on macro-texture (related to aggregate gradation, mix volumetrics and construction procedures), while with a wet surface, and at high speeds, safety mainly depends on macro-texture and air voids that rule characteristics such as friction and chemical resistance.

Potholes, bottom-up cracking, top-down cracking, and other surface distress impact pavement integrity, accident risk, tyre disruption, and vehicle costs. Mitigation measures include monitoring and maintenance [75-80].

To this end, it is noted that the same maintenance and rehabilitation processes involve risks because of work zones (which affect road safety and user costs) and because of virgin materials (which affect environmental risks). The mitigation of these risks can be carried out through the improvement of pavement design and the improvement of maintenance and rehabilitation using selective induction heating. This latter consists in using electromagnetic radiations to selectively heat a doped target layer [73], taking advantage of the self-healing of asphaltic materials [81].

Another source of risk is the excess of noise and loudness, with consequences in terms of human health and Disability-adjusted life years (DALYs [82],). To this end, strategies to reduce these risks include low-noise pavements and noise barriers.

Risks can increase because of natural and anthropic causes. On the one hand, the action of the atmospheric agents and the climatic changes (environment) must be considered during the design and maintenance process of transportation infrastructures [49]. On the other hand, ordinary and hazmat transport can affect performance and properties (e.g., pavement wear, and chemical resistance to hazmat spillage [42, 50];). Note that roadside hazards can be added to the causes described above since they can be both natural and anthropic. Examples of roadside hazards are [51]: i) Materials fallen from vehicles and abandoned/damaged vehicles; ii) Dead animals, stray livestock, and intruding shrubs or grasses; iii) Accumulation of objects and materials (ponding of water, trees, oil spills) on the traffic lane sides; iv) Damaged /missing drainage lids, surrounds and grates; v) Missing, illegible/damaged safety signs, safety barriers, guard rails and fencing at critical locations.

Furthermore, roadway assets and networks are vulnerable also to dangerous events such as earthquakes and floods. Different methodologies (e.g., physical damage assessment methods that are based on fragility functions and vulnerability indexes, traffic disruption assessment methods that use accessibility and link importance indexes) can be used to assess the seismic vulnerability of road assets and networks [52]. Because of climate change, flood risk is increasing [53]. Usually, historical data about atmospheric events are used to estimate flood risk, and this approach may lead to errors. Simulations (based on damage pattern, flood characteristics, precipitation depth, duration, number of cycles and pavement structure designs) and impact factor analysis can be used to try to reduce the aforementioned errors [53]. Results showed that (i) short-duration exposure to extreme precipitation should not cause significant damage, (ii) extreme events may affect the International Roughness Index (IRI) of a pavement; (iii) the pavement damage ratio can increase if the number of event cycles increases, (iv) long duration and higher cycle extreme events can effectively reduce the road pavement expected life, and (v) the Mechanistic-Empirical Pavement Design Guide (MEPDG) should be improved to take into account natural hazards and moisture damage of saturated old and new hot mix asphalt (HMA).

Another source of risk is soil erosion in cut, fill, and cut-and-fill road sections, especially in under resourced-countries. Seutloali and Beckedahl [54] considered this source of risk as critical for comprehensive land management decisions and monitoring strategies and underlined the need for further applications of assessment methods and control measures. In the latter study, it was highlighted how road construction mainly affects hydrologic and geomorphic processes because of the fact that it acts on (1) hill-slope profile (modification of natural profile, and formation of steep slopes, which are susceptible to severe erosion), and (2) vegetation cover (removal). In particular, road cut and fill embankments with steep slopes that are covered with insufficient vegetation, the interception of subsurface flows, and the concentration of runoff from the road surface can dramatically alter the natural equilibrium. Erosion control measures (e.g., field measurements, soil erosion control techniques, or soil erosion prediction models) can be used to mitigate this risk. For example, soil erosion control techniques that promote revegetation have the potential to reduce runoff (both velocity and quantity) and soil loss. Unfortunately, (i) there is a lack of methods that are efficient (from a cost point of view) and versatile (i.e., operational across landscapes that are different in type and size), and (ii) although sophisticated monitoring techniques are available (e.g., remote sensing technologies for erosion mapping), resource scarcity for assessing large areas limits their applicability.

Road-related risks are strictly connected to safety barrier issues ([43, 44], Road barriers: DM 223/92 and subsequent documents [83], Road barriers: EN 1317, parts 1-4 [84-87]).

Grzebieta [43] focused on the most dangerous crashes: (1) side impacts into narrow objects (@ impact speeds greater than 40 km/h); (2) head-on and large engagement offset crashes (@ about 120 km/h); and (3) roll-over crashes. If rollover does not occur, properly designed roadside or median barriers (rigid, semi-rigid, flexible, or temporary) can be used to safely redirect the vehicle, so that airbag and seat belt pre-tensioning systems do not activate. Butāns [44] pointed out that proper criteria should be used for the selection of a safety barrier. Based on standards and results from crash tests and simulations [43, 44], it is possible to state that both vehicle and barrier must be designed considering crashworthiness characteristics, such as (1) the ability to quickly (over a short distance) redirect and/or decelerate a vehicle respecting human tolerance/comfort levels (avoiding vehicle roll-over); (2) the barrier “containment level” (Low angle containment, Normal containment, Higher containment, Very high containment [88]); (3) the barrier’s “impact severity level”, expressed using the Acceleration Severity Index (ASI) and the Theoretical Head Impact Velocity (THIV) index, and classified into three levels (from A=high safety to B and C= significant passenger injury [88]); (4) the barrier “deformations” expressed by the working width (i.e., the movement sideways of the barrier after the vehicle impacts, which usually ranges from zero to few centimetres for rigid barriers, and from zero to 1-2 meters for the flexible ones. The European Standard EN 1317 [88] defines the levels from W1 to W8, where W1≤0.6 m, and W8≤3.5 m, the dynamic deflection (D), and the vehicle overhang (O); and (5) the timing of the vehicle airbag triggering (late or unnecessary airbag activation must be avoided).

Road markings and signs affect road safety and accident risk [89, 90]. Quality control and maintenance can reduce the risks deriving from the decay of their characteristics over time.

Note that, apart from road materials [91], the geometric design of a road affects road safety and accident ratios [38] in both rural and urban contexts. Unfortunately, these factors are difficult to change unless rehabilitation is carried out. Importantly, road safety audits and road safety audit reviews are tools that allow associating a score to a road [92]. This is an important point to consider when setting up a model for risk assessment.

Bridge approach settlement is considered by Helwany [45] as the cause of unsafe driving conditions, rider discomfort, poor public perception of the state infrastructure, structural failure of bridges, and long-term maintenance costs. Bridge approach settlements mainly depend on the poor performance of pavement, bridge abutment, consolidation of the backfill materials, consolidation of the foundation’s soils, and poor drainage. Possible mitigation methods are: (1) specifying more stringent backfill materials (granular foundation soils should be used instead of cohesive foundation soils [93],); (2) specifying compaction requirements (erosion and/or movement of backfill material can be avoided acting on embankment side slopes); (3) providing proper drainage; (4) use geosynthetic reinforced fill; and (5) use flowable fill controlled low strength material (which can be more expensive than geosynthetic reinforced fill for small quantity jobs).

At the same time, bump repair techniques include (i) asphalt patching or overlays, (ii) slab jacking for Portland cement concrete pavements, and (iii) replacement of the approach slab. In addition, it is recommended (1) to carry out inspections within six months of bridge completion, and (2) to use inclinometers to measure the lateral and vertical deflection of the backfill and foundation soils over time, as well as of embankment slopes and the related toes.

A further cause of road risk increase is related to bridge collapses. Deng [46] grouped the common factors resulting in a bridge collapse into two main categories: (1) Natural factors (e.g., flood, scour, earthquake, landslide, debris flow, hurricane, typhoon, and wind) and (2) Human factors (e.g., imperfect design and construction method, collision, vehicle overloading, fire, terrorist attack, lack of inspection and maintenance, etc.). In the same study cited above, collapse modes (for beam bridges, masonry arch bridges, steel arch bridges, and steel truss bridges) are reviewed and some mitigation measures are discussed. In more detail: (i) beam bridges need a proper design to avoid problems such as misalignment and failures; (ii) masonry and steel arch bridges should be designed to balance the in-plane and out-of-plane rigidities; (iii) steel truss bridges should have high levels of redundancy, and regular inspection and maintenance are needed to ensure high safety and serviceability levels; and (iv) flexible long-span bridges should be designed paying special attention on static stability, aerostatic stability (i.e., torsional divergence and lateral-torsional buckling), aerodynamic stability (i.e., flutter, buffeting, and vortex-induced vibration), stiffness and profile. If the aerostatic and the aerodynamic stabilities are not satisfied, considerable displacements or stresses can occur. To avoid wind-induced problems, the optimization of stiffness and profile is needed. Bridge-related risks are (1) Scour, caused by fast-flowing water (more attention should be paid to the scour depth prediction because of the fact that the available empirical equations, models, and technologies lack accuracy; real-time monitoring systems are needed [94]); (2) Bridge collisions (investigations on the bridge response to vehicles/ vessels collisions are needed); (3) Overloading (new intelligent systems can be useful to define loading-carrying capacity and restrict vehicle weight); (4) Wave loads during natural disasters (especially for coastal bridges during hurricanes and/or for bridges during earthquakes, where methods and formulas are able to estimate wave-structure interaction and to predict wave loads on bridge components are needed); (5) harmful vibrations caused by wind. Indeed, fluid-structure interactions are complex and additional effort is needed on (i) the application of new light materials, (ii) the extension of bridge span length, and (iii) the definition of the most important parameters for wind-resistant design. Finally, for the stability of arch masonry bridges, further studies on dilatancy accompanying sliding between adjacent blocks are needed.

The structural integrity of bridges, embankments and pavements is critical also from another standpoint. Indeed, it is strictly related to homeland security and the dual use of infrastructures [95]. Indeed, under the EU Regulation above, road surfaces should have the capacity to withstand “movements or transport of the overweight military assets on an occasional basis” (road/bridge gross weights of 130 t). Bridge joints are another source of risk for transportation infrastructures because they may undergo different types of failure [96]. Strategies to mitigate the associated risks include design and maintenance.

Many sources of risks relate to tunnels [97], including structural and functional problems, failure of lighting, fire-related risks [98], issues related to air filtration plants, and terroristattacks.

Road lighting is another source of risk and mitigation (cf. Road lighting: EN 13201, parts 1-5 [99]. Road lighting: EN 12665 [100]. Road lighting: Interegcentral Europe Dynamic Light (Interreg Central Europe)).

Events such as earthquakes, floods, traffic accidents, adverse weather, and human actions (e.g., poor maintenance interventions or industrial activities) affect the level of degradation of the transportation systems, and, for this reason, the term Degradable Transportation Systems (DTS) [12] was suggested. Based on this concept, an integrated equilibrium model, to estimate network reliability as a function of elastic traffic demand, independent component states (e.g., road capacity), and different levels of degradation, was proposed [12]. Note that, usually [12]: 1) Traffic demand is modelled as a constant variable; 2) Levels of degradation are assumed to vary between “full capacity” and “zero capacity”. 3) The aforementioned model considers the socio-economic impact of the levels of degradation of a system’s component, which grows when the time for repair/replacement increases [101].

Note that the degradability above pertains also to the obsolescence of the infrastructure and its ICT parts [102]. Traffic flow is crucial for both safety and security. It should be noted that to solve the traffic equilibrium problem, the following concepts should be considered. The degradation affects the capacity of the network (capacity degradation) and, consequently, its reliability, even though each network has a “reserve capacity” that should contain the fluctuation of the travel demand [12]. Consequently, the probability that a given traffic demand can be put up (at a sufficient level of service) by a network is defined as “capacity reliability” [12]. The combination of capacity reliability and travel time could be used as a comprehensive performance measure of the road network [12]. Other sources of risks refer to the use of electronics and telecommunication in smart and current roads [84-86].

For the use of optical fibres in transportation infrastructures and smart city applications, note that, apart from their use as a medium for telecommunication and computer networking, they can be used as sensors because of the dependency of several requested parameters (e.g., strains, temperatures, pressures) on the characteristics of the transmitted signals (e.g., intensity, phase, polarization, wavelength, and transit time). Consequently, they may be used also for safety-related purposes (e.g., accident detection and inappropriate position/speed of cars and pedestrians), traffic engineering purposes (e.g., speeds, congestions, times, vehicle count), and infrastructure-related issues (e.g., rockfall, landslide, and falling trees). Consequently, their failures (e.g., facet oxidation, contact degradation, and crystal grown-in defect) can involve massive risks for transportation systems and infrastructures (e.g., increased accident ratio, loss of information about the structural health status, loss of information about the load distributions). When it comes to setting up specific mitigation strategies, it is noted that these tasks are crucial but are very complex [103]. Furthermore, they involve many interacting technologies and fields (e.g., optical fibres, materials, structures, and transportation [104],).

For the use of radio frequency identification, RFID [105], note that RFID-based systems may consist of a laptop, a reader, an antenna, and an RFID, where the connections between the elements above are cabled with the exception of the last one (antenna to RFID tag). In turn, the RFID consists of a microchip and an integrated antenna. They can provide information about materials transportation, transportation security, and cracks [106]. At the same time, failures in crack monitoring could imply delays in pavement management. This type of risk appears quite high because the reliability of RFID tags is impaired by the same stresses that asphalt concretes undergo. To this end, increasing the number of tags and the frequency of readings can minimise this class of risk.

For strain gauges (electric or optical), they measure strains and their malfunctioning may affect monitoring, maintenance, and rehabilitation, with consequences in terms of risk of accidents. Here, possible strategies to reduce the risks include an increase in the number of sensors and the improvement of the supply chain (sensor, data log, energy supply, telecommunication). This applies also to Brillouin distributed fibre sensors, where continuous, real-time measurements along the entire length of a fibre optic cable are carried out [107].

This applies also to load cells, weight-in-motion devices, and piezoelectric devices, used to control how loads are distributed along the carriageway. Their use is important to assess the traffic spectrum in the pursuit of infrastructure management and security controls.

Importantly, accelerometers and microphones, used in monitoring infrastructure response (e.g., resonant frequencies), may suffer malfunctioning or failures and this impacts the promptness of maintenance and other strategies (e.g., road or bridge closure).

Other components that impact risk and reliability are cameras and traffic lights, with consequences on traffic management and infrastructure security and management.

When it comes to data transmission, network failures may impact risks in terms of accidents and infrastructure monitoring.

Air quality monitoring, especially in the urban context, is becoming more and more important, where, for example, the particles generated in the pavement-tyre interface are re-suspended and impact bystanders’ health (e.g., risk of respiratory infections). Sensors can detect Carbon dioxide (CO₂), Carbon monoxide (CO), NO, Nitrogen Dioxide (NO₂), and particulate matter (PM). These indicators help understand the quality of air. For example, PM concentrations are affected by pavement-tyre interaction (turbulence and resuspension) and impact human health. Failure in detecting and monitoring these indicators can increase the risk of inappropriate response to high PM levels, where pavement- and traffic-related mitigation strategies are supposed to be implemented [108]. Importantly, chemical detection is of paramount importance also by referring to threats (terrorism, weapons of mass destruction). Toxic chemical substances can be detected through wireless sensors and sensor networks [109].

Similarly, road weather information systems (RWIS) consist of 1) several environmental sensor stations (ESS), each one including sensors (e.g., thermometers, vanes, anemometers, udometers, air humidity sensors, pavement and subsurface temperature sensors, sensors for assessing pavement dry/wet/frozen state, sensors for assessing the amount of deicing chemical on the roadway and the freezing point of the road service, cameras), and remote processing units, RPUs. 2) Central processing units, CPUs. Indeed, RPUs have limited processing power and the data are sent from RPUs to the CPU. This latter is usually located close to a highway maintenance facility. CPU includes a database and other applications to collect, disseminate, and archive RWIS data [110]. 3) user interfaces (connected to the CPU). 4) other sources of information, including forecasts. Many opportunities, challenges and risks relate to RWISs, because of their ability to predict (weather and conditions) and consequently help manage (e.g., traffic, maintenance, work zones, and rehabilitation processes), inherent risks refer to wrong predictions due to unsatisfactory data processing or/and biased data. By referring to cameras, it is noted that their use can support road maintenance, crisis management, and homeland security (criminal law enforcement).

Note that poles (or smart poles) are part of smart cities/roads and can provide WiFi hotspots for different distances, smart lights, weather stations, cameras, and EV charging stations. The risks associated are the ones discussed above for each corresponding item.

Electrified roads (either overhead wiring or road-embedded power lines) are currently quite costly but they could allow electrified transport among most adjacent major cities [111]. Their interaction with pavement structure poses opportunities and challenges [112] but risks are quite high in terms of decay of the overall structural health status of the pavement (durability) and proper technologies to maintain, preserve, and rehabilitate are needed. Electrified roads are an emerging technology and standardizations and regulations are still missing. Consequently, matters such as security should be properly addressed [113].

On-board sensors (or in-vehicle sensors) are not included in transportation structures but they affect the overall performance of a transport system. They refer also to onboard diagnostics, Anti-lock Braking Systems, emissions monitoring, pavement monitoring (accelerometers, in-tyre sensors [114], support to automated driving, short- (ultrasonic, infrared, and capacitive sensors) and long-range (scanners and computer vision technologies) sensors for the representation of vehicle surroundings [115]. To this end, note that the knowledge of vehicle surroundings can come from different sources, such as onboard sensors (as mentioned above), sensors in the infrastructure, and communications (V2X, I2X). Failures of the systems above affect road safety. Mitigation measures are quite difficult to set up but the same V2V communication systems could help improve the single-vehicle representation of vehicle surroundings.

In the new context of the massive use of sensors for multiple purposes, the reliability of transportation infrastructures could depend on the reliability of these systems and namely on the failure probability of computing and communication processes (e.g., infrastructure to everything, I2X, vehicle to everything, V2X, and pedestrian to everything, P2X, communications [116, 117]. This issue affects infrastructure integrity, risk, resilience, autonomous vehicles, road safety, and the same use of transportation infrastructures as lifeline infrastructure where the failure of these lifeline services during natural hazard events could impact populations by exacerbating the hazard itself and/or hindering their ability to respond to or recover from the event [118].