Road Safety Manual
A manual for practitioners and decision makers
on implementing safe system infrastructure!

You are here

4.6 Safe System – Scientific Safety Principles and their Application

The Safe System approach marks a shift from a sole focus on crash reduction to the elimination of death and serious injury. Well-established safety principles underpin the Safe System approach as set out in Key Developments. Further principles include the following:

  • The combination of infrastructure safety features, vehicle safety features and travel speed(s) of crash-involved vehicles determines the impact forces that humans are subjected to in any crash. These interactions are to be managed to avoid fatal or serious injury outcomes.
  • Safety levels are to be the key determinant of sustainable mobility levels. Travel speeds require management to target levels below those known speed thresholds that deliver fatal or serious injury crash impact energies (based on the level of vehicle safety and mix, and the nature of protective infrastructure characteristics; See also Safe System Principles and Safe System Elements).
  • Safe System approaches typically aim to develop a road transport system that is better able to accommodate human error by providing a safe operating environment - despite human fallibility - and providing effective post-crash care.
  • A system-wide intervention strategy addressing all crash-phases and all Safe System elements is to be adopted, which addresses the safety of all road users.
  • Legislative and enforcement strategies that achieve widespread user compliance with road rules and laws are necessary, as are strategies which deter the entry and exit arrangements of users and vehicles to the road system.

Development of the Safe System Approach

As noted earlier, the Safe System approach builds upon the ground-breaking road safety efforts of the Netherlands and Sweden.

Netherlands sustainable safety

Wegman & Aarts (2006) outlines a set of guiding principles (based on the Dutch Sustainable Safety Vision) considered necessary to achieve sustainably safe road traffic. The principles are based on scientific theories and research methods arising from disciplines including psychology, biomechanics and traffic engineering, and are set out in Table 4.2 below.

Table 4.2: Sustainable safety principles
Sustainable Safety PrincipleDescription

Functionality of roads

Single function of roads as either through roads, distributor roads, or access roads, in a hierarchically structured road network.

Homogeneity of mass and/or speed and direction

Equality in speed, direction, and mass at medium and high speeds.

Predictability of road course and road user behaviour by a recognisable road design

Road environment and road user behaviour that support road user expectations through consistency and continuity in road design.

Forgiveness of the environment and road users

Injury limitation through a forgiving road environment and anticipation of road user behaviour.

State of awareness by the road user

Ability to assess one’s task capability to handle the driving task.

Sweden’s vision zero

Tingvall (2012) commented on the challenges Sweden faces in redefining transport policy principles to reflect Vision Zero (or the Safe System approach):

  • Sweden has found a new way in recent years to express transport policy. In summary, ‘accessibility can only be developed within the framework of safety and environment’.
  • While mobility and accessibility make up the functionality of the transport system, the safety parameter (like many other elements in a society) is not a variable in an equation, but has threshold limits that cannot be exceeded.
  • This shift is gradual. The new Swedish speed limit system is a good demonstration of how mindsets shift over time, with 80 km/h now the maximum speed for an undivided road (unless there are low traffic volumes), and this is intended to be absolute.
  • While obtaining public acceptance is challenging, Sweden is gradually changing its approach to permissible travel speeds, and therefore to the setting of speed limits on new roads, or determining the investments necessary to modify existing roads to allow higher limits, based on the new speed limit system. This is where the safety benefits will be realised.
  • The mobility needs will then be the deciding parameter for the infrastructure investment necessary to have safe higher travel speeds. Is it worthwhile to invest in safety solutions to increase mobility? This is the right question, but it has taken many years to have this rather natural logic understood and accepted (See Box 4.2 below).

Box 4.2: Fundamental shift in thinking: Safety as the limitation mobility – inherent to a Safe System approach

  • Traditional approach – road authorities have striven to specify desired mobility (travel speeds) while doing what they can to improve safety on a length of road.

“You can travel from A to B at 100 km/h and we will make some improvements to this two lane two way rural road to improve your travel safety:

  • Safe system thinking – achieve safe travel, by determining the travel speed which can be adopted without risk of death or serious injury on this length of road.

“You can travel at this safe speed from A to B based on the safe system elements which are operating and which will avoid fatal and serious injury in the event of a crash. You can only travel faster if infrastructure safety is improved.” (eg, roundabouts at intersections, median barriers, run off road barriers to protect from roadside objects, etc.)

Understanding the Safe System Model

The elements of the integrated, human-centric Safe System model for safe road use and their interactions can be depicted as follows (Figure 4.1):

Figure 4.1: A model of the Safe System approach - Source: Adapted from OECD/ITF, 2008; ATC, 2009.

The Safe System design model has four main elements (including alert and compliant road users) plus five supporting activities that can be adjusted and applied in agreement with the four main elements to assist in making crashes more likely to be survivable.

The four main design elements are:

  • Safe roads and roadsides – that are predictable and forgiving of mistakes. They are self-explaining in that their design encourages safe travel speeds and help avoid errors.
  • Safe speeds – travel speeds that suit the function and level of safety of the road. People understand and comply with the speed limits and drive to the conditions.
  • Safe vehicles – that prevent crashes and protect road users, including occupants, pedestrians and cyclists, in the event of a crash.
  • Safe road users – road users that are alert and unimpaired, and who comply with road rules. They take steps to improve safety, and demand and expect safety improvements.

The key supporting Safe System elements include:

  • emergency medical management for post-crash care (the fifth UN Decade of Action pillar);
  • understanding of crashes on the network, which requires good data to enable risks across sections of the network to be accurately identified;
  • control of admittance (entry and exit) of drivers to/from the road transport system (licensing arrangements including graduated licensing arrangements);
  • effective legislation and systems, enforcement and justice system support;
  • educating and informing the public.

The last three elements in the list above support achieving road user compliance with the road rules.

In summary, for alert and compliant road users, a combination of vehicle safety features, safety characteristics of the infrastructure and travel speed are required, together with effective emergency medical post-crash care, in order to avoid a fatal or disabling serious injury outcome in the event of a crash.

The Critical Role of Travel Speed in Achieving a Safe System

Travel speeds are a critical variable within a Safe System with allowable safe speeds on any part of the network being dependent upon vehicle types (and their protective features), the forgiving and protective nature of the infrastructure and roadsides, the restrictions upon roadside access to the roadway and the presence of vulnerable road users. All of these factors will determine a maximum vehicle speed on each section of the network above which an unacceptable probability of death is likely from any collision.

Injury outcomes and the creatioin of an inherently safe road system are largely dependent on the kinetic energy in the system. At the moment of impact, the force on the boady can exceed the body's tolernace.

The amount of kinetic energy (Ek) is calculated according to the following formula: 

Ek = 1/2 mv2

where m = mass (kg) and v = velocity (m/s)

The formula shows that kinetic energy does not increase linearly with speed, but with the square of the speed. This has important implications for how speed affects the road safety. A small incrase in travel speed can signficantly incrase the probability of a serious injury outcome. 

Source: Austroads (2018)

speed management is also critical to other societal goals

Reducing speed is one of the most effective ways to improve safety, saving lives and debilitating injuries. However, the opportunities that lowered travel speeds offer to other societal goals are generally under-appreciated. Reducing speed also generates multiple other benefits funamental to sustainable moblity: reduced climae change impacts of road trasnport, incrased efficiency (fuel and vehicle maintenance), improved inclusion and walkability (Job et. al. 2020)

Job et. al. (2020) highlighted that when taking into account the full range of economic impacts (GHGs, emissions, fuel, etc.), economically optimal speeds are lower than expected and typically lower than prevailing speed limits compared to limited evaluations of impacts based on travel time savings alone.

Speed management can be achieved through a range of interventions including road infrastructure and vehcile technology, as well as enformcement and promotion.

Mean speed and crash risk

Crash outcomes, especially fatal crash outcomes, are influenced directly by the travel speed of vehicles at the time of impact.

Elvik et al. (2004) report that “Speed has been found to have a very large effect on road safety, probably larger than any other known risk factor. Speed is a risk factor for absolutely all accidents, ranging from the smallest fender-bender (crash) to fatal accidents. The effect of speed is greater for serious injury accidents and fatal accidents than for property damage-only accidents. If government wants to develop a road transport system in which nobody is killed or permanently injured, speed is the most important factor to regulate”.

Table 4.3 from Elvik et al. (2004) sets out the effects of variations in mean speeds on crashes of various severities. This relative change relationship applies on all lengths of road, over comparable periods of time and refers to the effects of changes in mean speed of travel of all vehicles.

Table 4.3: Change in accidents resulting from a change in speed - Source: Adapted from Elvik et al. (2004)
 Relative change (%) in the number of accidents or victims
Change in speed (%)-15%-10%-5%+5%+10%+15%
Accident or injury severity      








Serious injuries







Slight injuries







All injured road users







Fatal accidents







Serious injury accidents







Slight injury accidents







All injury accidents







Property damage only accidents







Fatal crash outcomes are the crash type most affected by speed variation. As the above table shows, even small changes in speed (+5%) are associated with very large changes in the number of road crash fatalities (+25%).

Influencing kinetic energy level in crashes

As indicated in the safety principles above, an important way to reduce fatal or serious injury crash outcomes is through better management of crash energy, so that no individual road user is exposed to crash forces that are likely to result in death or serious injury.

Conditions that support limiting crash energy to levels below which fatal or serious injury crash outcomes are relatively unlikely, are now becoming better understood, but are still not well recognised or applied system-wide in most countries.

A key strategy is therefore to move (over time) to set posted speed limits in response to the level of protection offered by the existing (or improved) road infrastructure and the safety levels of the vehicles and vehicle mix in operation on sections of the network.

Mobility needs to be constrained by Safe System compliance. Future safe infrastructure investment will often be necessary before considering raising the speed limits on sections of the network in order to avoid increased fatalities or serious injuries.

Box 4.3: An alternative road infrastructure design approach

Tingvall (2005) notes that while infrastructure design has previously been built on crash prevention, the alternative Safe System philosophy is built on kinetic energy management and injury prevention (secondary prevention rather than primary prevention). Speed is now more related to the outcome of an incident or crash rather than the driver’s ability to keep the vehicle in control. This has led to more extensive use of roadside and median barriers, while intersections are redesigned to roundabouts, and roadside access (and development) needs to be more fully controlled. These are examples where the number of crashes might increase, but where the outcome is controlled so that human tolerance to serious health losses is not exceeded.


McInerney & Turner state that the discipline of managing energy exchange and related forces currently exists in the fields of structural engineering for buildings and mechanical engineering for machines, but is rarely sighted in the design of roads. For infrastructure to provide the key building blocks for a Safe System, road engineering design practice worldwide must include provision for the management of kinetic energy. For example, there are simulation programs that examine an errant vehicle departure into a roadside environment, which calculate the change in kinetic energy as the errant vehicle encounters roadside hazards. The rate of kinetic energy dissipation can then be translated into differing collision severity potential. 

Reference sources

No reference sources found.