9.2 General Principles of Infrastructure Safety Management

In general, it is believed that driver error contributes to a large proportion of road crashes, with some studies suggesting that human error has played a role in over 90% of crashes (e.g. Sabey, 1980; Treat, 1980). Although the role of human error in road crashes is substantial, such figures downplay the significant role that infrastructure can have in achieving Safe System outcomes (also see the discussion on this issue in The Safe System Approach).

Findings from Sweden identified that road-based factors were most strongly linked to a fatal crash outcome. Stigson et al. (2008) reviewed fatal crashes based on in-depth crash investigation, with crashes categorised based on factors that contributed to the crash outcome (as opposed to crash causation). The study identified that there were strong interactions between the three system components (vehicles, road infrastructure and road user issues), but that road-based factors were most strongly linked to a fatal crash outcome.

Further evidence of the role infrastructure plays in fatal and serious injury crash outcomes can be found from research that investigates the benefits of infrastructure safety treatments. Various studies have identified that well-designed infrastructure (such as roundabouts and protective barrier systems) can reduce fatal and serious injury crash outcomes by up to 80%. This reduction can occur regardless of whether crashes were the result of human error (also see Design for Road User Characteristics and Compliance). For further information on effective treatments, see Intervention Selection And Prioritisation.

There is a strong economic argument for the provision of safe infrastructure. Examples exist from many countries that demonstrate the benefits from targeted road safety improvements. OECD (2008) identified that such targeted improvements can deliver up to 60 times the benefit compared to the cost (i.e. for every $1 spent, benefits of up to $60 in terms of crash cost savings can be achieved). As identified by the UNRSC (2010), few other infrastructure investments produce the economic benefits of infrastructure targeted at improving road safety. Take for instance the installation of cable median barrier to reduce cross median crashes as this has become common practice in some nations (for instance Sweden and the United States). Even more substantial investment programmes are able to return substantial safety benefits when compared to their costs. An analysis undertaken by iRAP (www.irap.org) on improvements to road infrastructure on the worst 10% of roads (i.e. those roads with greatest number of death and serious injury) in each country identified substantial potential gains when comparing the costs with the benefits. The average over all countries was a benefit-cost ratio (BCR) of 8:1 (i.e. $8 worth of benefits for each $1 invested). This ranges from a BCR of 5:1 in high-income countries, to 19:1 in upper middle income countries over a 20 year period.

The European Commission, under the Horizons 2020, developed the SafetyCube project as an innovative road safety Decision Support System (DSS). The intent of this effort was to allow policy-makers and stakeholder the ability to select and implement the most appropriate strategies, measures and cost effective approaches to reduce casualties of all user types and all severities in Europe and worldwide (see: SafetyCube). The United States also has the Crash Modification Clearinghouse the includes over 7000 (2019) interventions and the potential crash reduction for each countermeasure (See: Crash Modification Clearinghouse). Both of these tools provide significant information and resources on the use of road infrastructure to reduce crash potential.

A solid understanding of key infrastructure principles is required by road agencies and others responsible for delivering road safety. Some of the key elements of relevance to the development of infrastructure policy, standards, guidelines and tools include:

• Understanding of, and adherence to, the Safe System approach, including acknowledgement that designers are ultimately responsible for the design, operation and use of the road system, and therefore the entire safety of that system (UNRSC, 2010).
• Safety needs to be embedded in planning and design. In order to have maximum and cost-effective impact, safety needs to be included as early as possible. This includes ensuring that new projects have safety embedded and that existing roads are upgraded to account for Safe System principles.
• Priority policy issues for major roads include the need for a clear distinction and separation between inter-urban high-speed roads and urban roads (i.e. hierarchy); strong control and management of land use and urban development; and planning cities and communities for those without cars (i.e. walking, cycling and public transport). (UNRSC, 2010).
• There is a need to identify high risk locations through the traditional approach involving analysis of crash locations, and a proactive approach based on design elements and road and roadside features. Combining these approaches will maximise opportunities to identify risk. The focus when identifying high risk locations should be primarily on fatal and serious injury.
• Risk locations should be addressed in a cost-effective manner. Again, the focus should be on the elimination of fatal and serious injury.
• There is a requirement for monitoring, analysis and evaluation of the network, including safety performance and the impact of changes that are made. This assessment of change is often overlooked but is required to ensure expected outcomes are met.
• Monitoring, analysis and evaluation allows changes to be made based on the outcome to risk reduction throughout the system. It is important to establish a method and a strategic approach about how to carry out evaluation of the system, and how changes to the safety system or program will consider addressing the new knowledge gained from the evaluation.

Guidance on the risk assessment process has been developed across many different industries and activities, including road safety. The process (briefly introduced in Figure 1) involves the identification of high risk locations; analysing data to determine the cause of this risk; identifying evidence-based solutions that are effective in addressing the risk; implementing these solutions; and then monitoring and evaluating the outcome. Each of these stages is explained in detail in Assessing Potential Risks And Identifying Issues to Monitoring, Analysis and Evaluation of Road Safety.

In broad risk assessment terms, the chance of sustaining death or serious injury can be decreased by reducing the:

• exposure to the risk (an example may be to divert traffic from low quality roads to higher quality ones), shortening crossing distances for pedestrians
• likelihood of the crash (this includes the provision of a predictable road environment), reducing conflicts between vehicles and vulnerable road users);
• severity of the crash (for example, by providing a forgiving roadside to reduce harm if a vehicle does leave the road, reducing conflicts between vehicles and vulnerable road users);

With an understanding of these factors, crashes can be influenced in a number of ways through changes in the road environment. As examples, improvement in safety can be gained from:

• reducing exposure;
• regulating/controlling movements and turns, especially at intersections and access points;
• reducing speeds;
• reducing conflict points;
• separating vehicles of different mass (e.g. specific facilities for pedestrians and cyclists) and travelling in different directions (e.g. median barriers);
• warning road users of unusual or risky features (e.g. providing advanced warning);
• providing adequate information to enable road users to negotiate the roadway safely;
• removing hazards (e.g. utility poles and trees from the roadside);
• protecting road users from hazards that cannot be removed (e.g. providing guardrail and median barriers);
• improving surface friction.

Engineering-based treatments generally work by influencing one or more of these factors. Examples of such treatments and their effectiveness can be found in Project-level and Network-level Approaches.

The Role of Policies, Standards, Guidelines and Tools

Standards, guidelines and tools are the mechanisms that support the consistent interpretation and delivery of policies. Policies set the framework for road safety activity, and without these, delivery of road safety is reactive and lacks structure. The policies will often set the direction at a high level and will also contain direction on how to achieve standards using a predetermined set of criteria. Guidance on strategy and policy development can be found in Road Safety Targets, Investment Strategies Plans and Projects and in Development of Policies, Standards and Guidelines in Policies, Standards and Guidelines. These Policies, standards and guidelines are often slow to change, either because of lack of research, unwillingness to change current practice or lack of knowledge related to the given system.

Changing established practice is often difficult, and careful management of this process is required. Strong leadership is needed to facilitate policy shift, and this needs to happen in parallel with an update of corresponding policies, standards and guidelines. Change requires understanding and awareness, the desire to create change, the knowledge to implement the change, and the ability to do so within the environment or system an individual works within.

Once policies are set, there is need for linkage to standards guidelines and how the associated criteria for meeting the standards can be achieved. Standards (as well as road rules and regulations) dictate those things that must be done to achieve a predetermined level of quality or attainment. In many countries standards also have a legal basis that are adopted in design and operational manuals. Criteria are the specifications for achievement of the standards and are typically detailed as policy. Guidelines provide direction on how things should be done but are not necessarily requirements. In contrast to established policy and standards, there is scope to deviate from the advice provided in guidelines, but this must be justified and assessed for the impact on safety outcomes. When deviating from policies and standards documentation outlining the reasoning is often used. Policies and standards are typically based on many years of experience and outcomes from research and understanding. Guidelines, because they are not requirements may contain reasoning for how the standards and policies were developed, how to apply them in different circumstances, and might provide ranges of values to consider based on the conditions encountered.

It is important to note that compliance with policy, standards and guidelines does not mean that safety will be maximised, nor minimised when they are not achieved. There are many examples where new roads have been built to standard but have a less than desired safety outcome. Policy, standards and guidelines are often dated, and may not include adequate content based on Safe System principles. The manuals that contain the policies, standards and guidelines, generally offer the minimum acceptable values for design. Departing from the minimum requirements sometime occurs when the financial, environmental, right of way, or social cost are high and the relative change in safety risk is low. These tradeoffs are often documented to discuss reasoning for not achieving the policy, standards or guidelines.

There is much to be gained by looking to other countries for guidance on setting new policies, standards and guidelines, and such an approach is important for benchmarking (also see Country Management System Framework in Management system Framework and Tools). However, often safety policies, standards and guidelines are directly copied from other countries without due consideration of local conditions, design for vulnerable road users, different vehicle types and road user compliance.

Also, in many circumstances, policies often provide fewer options for use in constrained environments. It is typical that several compromises need to be made in road design. When combined, these issues can lead to poor safety outcomes (also see the discussion in Global Level in Policies, Standards and Guidelines on minimum criteria, and the concept of extended design domain). Typically, an assessment of the likely road safety impact is required to ensure that safety objectives are met. It is for this reason that approaches such as road safety audit (see Road Safety Audit in Proactive Identification) are required, and that when undertaken, these are not just a check against standards and guidelines.

Knowledge of the safety implications of design decisions is constantly improving, and with this there is sometimes a need to update policies and procedures. This includes the need to update standards, guidelines and tools.

Box 9.1 indicates how a policy decision that was initially driven by economic reasons at the political level has resulted in safer road design principles on major roads in New Zealand.

Box 9.1: New Zealand review of design standards for Roads of National Significance

Linkage with other Policies, Standards and Guidelines

Delivery of road safety infrastructure does not occur in isolation, and it is important to consider broader safety, road management and societal issues when developing policies, standards and guidelines. Similarly, it is important to advocate for road safety outcomes when developing broader transport and related policies. The Belize case study provides a useful demonstration of linking infrastructure improvements to other safety improvements. Land use measures have a strong linkage to safety outcomes, an issue that is often overlooked in LMICs. Such measures define the type and intensity of the generated traffic, and the way it enters and exits the roadway. A detailed discussion on this issue can be found in Impacts of a Safe System Adoption on roles and Responsibilities of Authorities.

There are also links that can be made with broader policy agendas, as illustrated by the case study from the Netherlands.

As a further example, asset management involves maintaining and upgrading road infrastructure, and this typically has road safety implications. It is often the case that planning and funding for asset management and road safety outcomes occur in isolation, and without adequate linkages between the two and this can impact how a project is ultimately designed and constructed. Both activities are closely linked, with each directly influencing the other. Adequate knowledge of the safety implications of asset decisions is required when establishing policy and practice. Similarly, safety decisions can have a substantial impact on asset management (particularly costs for maintaining assets). For instance, the installation of unnecessary roadside crash barrier on flat slopes would increase the potential for crashes, increase maintenance and replacement costs, and not reduce the injury potential. When installed correctly the crash barrier systems can be used to shield other assets from impacts, close gaps in locations where injury potential is high and protect workers who may need to maintain the assets.

When considered in isolation, the two road management approaches are often thought to act in conflict. There may be a perception that increases in funding for road safety may mean less funding or increased expenditure for asset management. However, there is some clear evidence that the two can act in harmony to produce benefits that are greater than those that can be delivered when considered in isolation.

Specific examples from LMICs are scarce, but the example in Box 9.2 from Australia serves to illustrate the level of benefits that can be gained through a coordinated approach. Combining the safety benefits with those from asset improvements can often lead to better project viability. This issue is discussed further in Roles, Responsibilities, Policy Development and Programmes.

Box 9.2: Asset management and safety objectives can be complementary