Article Text

Download PDFPDF

Evolution of a holistic systems approach to planning and managing road safety: the Victorian case study, 1970–2015
  1. Carlyn Muir1,
  2. Ian R Johnston2,
  3. Eric Howard1
  1. 1 Monash University Accident Research Centre, Monash University, Clayton, Victoria, Australia
  2. 2 Accident Research Centre, Ian Johnston Transport Safety, Glen Waverley, Victoria, Australia
  1. Correspondence to Professor Ian R Johnston, Accident Research Centre, Ian Johnston Transport Safety, Glen Waverley, Victoria 3150, Australia; irj111{at}


Background The Victorian Safe System approach to road safety slowly evolved from a combination of the Swedish Vision Zero philosophy and the Sustainable Safety model developed by the Dutch. The Safe System approach reframes the way in which road safety is viewed and managed.

Methods This paper presents a case study of the institutional change required to underpin the transformation to a holistic approach to planning and managing road safety in Victoria, Australia.

Results The adoption and implementation of a Safe System approach require strong institutional leadership and close cooperation among all the key agencies involved, and Victoria was fortunate in that it had a long history of strong interagency mechanisms in place. However, the challenges in the implementation of the Safe System strategy in Victoria are generally neither technical nor scientific; they are predominantly social and political. While many governments purport to develop strategies based on Safe System thinking, on-the-ground action still very much depends on what politicians perceive to be publicly acceptable, and Victoria is no exception.

Conclusions This is a case study of the complexity of institutional change and is presented in the hope that the lessons may prove useful for others seeking to adopt more holistic planning and management of road safety. There is still much work to be done in Victoria, but the institutional cultural shift has taken root. Ongoing efforts must be continued to achieve alert and compliant road users; however, major underpinning benefits will be achieved through focusing on road network safety improvements (achieving forgiving infrastructure, such as wire rope barriers) in conjunction with reviews of posted speed limits (to be set in response to the level of protection offered by the road infrastructure) and by the progressive introduction into the fleet of modern vehicle safety features.

  • policy

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.


There have been over 189 000 deaths on Australian roads since 1925, when record keeping began.1 In the last four and a half decades, not only have per capita and per exposure fatality rates continued to fall but the absolute number of deaths has fallen by around two-thirds.1 While this pattern of more or less continuous improvement is common to many Western motorised nations, the rates of gain differ considerably across the motorised world.

In Australia’s federal system, government responsibilities for road safety vary across jurisdictions. The Australian Federal Government is responsible for regulating safety standards for new vehicles, and for allocating infrastructure resources (including safety), across the national highway and local road networks. State and territory governments are responsible for funding, planning, designing and operating the State or Territory’s road network; managing vehicle registration and driver licensing systems; and regulating, enforcing and educating road user behaviour. Local governments have responsibilities for funding, planning, designing and operating the road networks in their local areas.1

Victoria is Australia’s geographically smallest state, at almost 240 000 km2, yet the most densely populated with 5.9 million people.2 In Victoria, road deaths reached a historical high in 1970, when the absolute number exceeded 1000 for the first time.3 This resulted in a watershed public demand for change (evidenced by an ongoing campaign by the major daily newspaper to ‘Declare War on 1034’), with the first outcome being seatbelt-wearing legislation, which represented the first mandatory seatbelt law in the world.4 By early 1972, similar legislation had been enacted in all six Australian states.3 Physicians, and other road safety advocates, continued to push for further road safety legislation in the 1980s and 1990s, and a range of other interventions were introduced, including random breath testing (RBT); a zero blood alcohol concentration (BAC) for learner, probationary, heavy vehicle and public transport drivers; mandatory use of safety helmets for bicyclists; and automated (camera) enforcement of speeding.3

Rather than embrace the holistic Safe System model, Victoria was a leader in regulating and enforcing road user behaviour and supporting this approach with public education. Despite excellent advances in traffic safety science over this time, the major focus of public policy remained determinedly on behaviour change and the search for ‘perfect’ road user behaviour, especially around the risk factors of drink-driving, speeding and ‘aggression’.3 While many of the adopted interventions demonstrated impact, researchers began to identify design and operational problems in the system that influenced road user behaviour, at odds with the public policy position that laid blame solely at the door of the users.5–8

Among traffic safety professionals, the individual road user came to be regarded less as ‘the predominant cause’ and more as the system’s weakest link, with the focus shifting to a consideration of how road and traffic system design might seek to accommodate common human errors.3 Key institutional leaders have come on board, but shifting politicians and the community to a similar understanding remains a work in progress, although some gains have been made.

Vision Zero

The Vision Zero philosophy aimed to achieve a road transport system with no fatalities or serious injuries, and in September 1997 it was enshrined in Swedish law.9 There was strong support for Vision Zero among Australian researchers, driven largely by the appointment, in the late 1990s, of one of the Swedish architects of Vision Zero to Victoria’s key road safety research centre. However, debate ensued regarding the feasibility of a goal of zero serious injuries in Australia.5 Transport system engineers, in particular, were concerned about the cost and feasibility of achieving zero deaths, and continued for a long time to argue for a trade-off to maximise the economic benefits of mobility while managing the costs of death and serious injury.5 Over time, it became clear that Australian governments would not fully support Vision Zero in the way it had been adopted in Sweden. For example, the National Road Safety Strategy 2011–20201 set a target to reduce death and serious injury by 30% by 2020. Even if this target were achieved in the first year (rather than the targeted year 10), this is tantamount to accepting that over 200,000 Australians would be seriously injured or killed during the life of the strategy. Safety was only one goal in a trade-off process, and so long as human behaviour continued to be seen as the primary cause of crashes, ambition would remain low.

Safe System approach

A less ‘radical’ approach to change was proposed for Victoria, which emerged from a combination of the Vision Zero philosophy and the Sustainable Safety model developed by the Dutch.5 The Safe System approach reframes the way in which road safety is viewed and managed. Central to this approach is the proposition that road trauma cannot be traded for improved mobility. It aims to address all elements of the road transport system in an integrated way to ensure that the kinetic energy levels in crashes remain below what the human body can tolerate10 (figure 1).

Figure 1

Safe System approach.30

Accordingly, there is a need to address each of the key elements of the transport system—the road and roadside, the vehicle and road user behaviour—as well as review the way in which safety is managed and coordinated across agencies and groups.11 It recognises that the responsibility for safe operation of the network is shared between many individuals and organisations, including those who construct and maintain the roads, set the speed limits, make the laws, provide the vehicles, make the land use planning decisions affecting traffic flows and roadside access, use the network, enter contracts for supply of transport services, enforce compliance, employ drivers to use the road network in their work, operate the emergency health system and more.3

In the Safe System framework, the individual road user remains a critical part—but a part only—of shared responsibility; each user has a responsibility to obey all traffic laws and to ensure fitness to drive by avoiding impairing substances and managing fatigue. This changed view of road user responsibilities away from solely a ‘blame the road user’ emphasis is a key feature of a Safe System and the most difficult conceptual shift for governments, creating an important set of challenges in determining and monitoring ongoing performance against the accountabilities of all players.3

Victoria’s introduction to Safe System

The Safe System conceptual approach was notionally adopted in Victoria in 2004 before later being endorsed by the Australian Transport Council to underpin the national 2004–2005 Road Safety Strategy. This signalled an acceptance of the need for a paradigm shift in road safety thinking, research and strategy.5

However, achieving the shift is painfully slow. For some years, there was more change in the language among policy makers than in action on the ground. Nevertheless, several programmes were introduced to enhance the safety of roads, roadsides and vehicles, with a boost in funding to address the largest infrastructure contributions to road injury.5 In 2004, Victoria commenced a Safer Road Infrastructure Program, initially investing $A130 million for 113 projects designed to address the major crash-type risks across the Victorian road network (encompassing the system of interconnected roads designed to accommodate vehicles and pedestrian traffic). A further investment of $A110 million for 252 projects was allocated in 2006, followed by a commitment of $A650 million in 2007 for projects to be carried out over the subsequent 10 years.5 ,12 These are unprecedented levels of infrastructure safety investment, and are important markers of a shift in deed as well as in word.

Vehicle technologies have also received attention. For example, lane departure warning technology requires edge-lining, and this is taking place progressively; crash avoidance technologies, electronic stability control (ESC) and in-vehicle speed warnings are being promoted and are slowly penetrating the market; and Automatic Braking Systems (ABS) for motorcycles is being promoted voluntarily.

A conceptual shift in approach requires close cooperation among all the key agencies. Victoria was fortunate in that it had a history of interagency cooperation. The road and traffic agency, the injury compensation insurer, the Police, the Justice Department and health agencies had long met regularly to coordinate their efforts, and reported collectively to their Ministers of State, who in turn reported collectively to Parliament.

The monopoly (government-controlled) injury compensation insurer played a pivotal role by realising that an investment in injury prevention resulted in reduced financial liabilities.13 This agency invested heavily in public education to support legislative and enforcement efforts and, importantly for a paradigm shift, directly invested in safer infrastructure programmes. A paradigm shift in thinking is a key first step, but achieving implementation in practice is quite another.

Safe System challenges in Victoria

Safe System implementation in Victoria has not been a straightforward journey. The challenges are generally neither technical nor scientific; they are predominantly political and social.5 Partly, this is a result of a fundamental misunderstanding of what a Safe System seeks to achieve. Safe System has three elements: it is a vision, it is a set of principles and it is a set of tools (summarised in box 1).

Box 1

Safe System vision, principles and tools


  • Safe System thinking involves a holistic view of the road transport system and the interactions among roads and roadsides, travel speeds, vehicles and road users. It is an inclusive approach that caters to all groups using the road system and recognises that people will always make mistakes and may have road crashes—but the system should be forgiving of errors and those crashes should not result in death or serious injury.


  • No death or serious injury occurring on our road network is acceptable.

  • The health and well-being of our society should not be traded off against other societal benefits.

  • Humans are fallible and prone to error.

  • Humans have tolerance thresholds to energy exchange, above which the absorption of energy is likely to lead to death or serious injury.

  • Road users should not be subject to energy exchanges that will lead to the likelihood of either death or life-disabling injury.


  • In order to ensure that Safe System elements are considered, or to measure how well a given project (eg, an intersection, road length, area, treatment type and others) aligns with Safe System principles, a Safe System matrix has been produced by Austroads.31

  • The purpose of the matrix is to assess different major crash types (those identified as the predominant contributors to fatal and serious crash outcomes) against the exposure to that crash risk, the likelihood of it occurring and the severity of the crash should it occur.

  • The Safe System assessment framework is best applied by teams of road practitioners with varied types and levels of experience.

Community support for Safe System

Achieving community support and understanding about the purpose of Safe System in Victoria is ongoing, and is fundamental to overall success. It is instructive to recall that Victoria’s seminal intervention (legislating seatbelt use) stemmed directly from public outcry that followed the 1970 high road fatalities. The ‘Declare War on 1034’ campaign helped build a climate of concern which contributed, in turn, to the passage of the world’s first mandatory seatbelt-wearing law, a law with a dramatic immediate effect and a global legacy.3 4 14 This resulted in strong commitment across the political spectrum (while not necessarily supported by the political leadership on both sides) and reinforced by highly supportive commercial print media. Without public acceptance of the need for urgent action, the seatbelt-wearing legislation would never have passed. However, it is important to note that this pioneering action was predicated on a perceived need to change road user behaviour, not to develop a safer road transport system as a whole.

Stakeholder positions

The application of Safe System thinking requires all the actors in the system to interpret the principles in the same way, and to plan actions that are consistent with the philosophy.5 However, the action plans that follow from those strategies still very much depend on what politicians perceive to be publicly acceptable. They remain, for the most part, demonstrably reactive,3 as summarised by Professor Fred Wegman—a key figure in the Dutch approach to the adoption of a holistic approach to road safety planning and implementation: ‘While the Safe System concept has been present in Australia for many years, its implementation still proves a challenge to everyone involved in road safety’. 15 Professor Wegman’s observation is very astute—Australian jurisdictions (including Victoria) have embraced the language far more than the action. Tackling bad behaviour remains perilously popular; spending scarce resources on system-wide change is not. Table 1 outlines the positions on Safe System thinking of the various key stakeholders in Victoria.

Table 1

Positions on Safe System thinking of Victorian key stakeholders

Infrastructure investment

A characteristic of road transport in Australia (including Victoria) is the large scale of the country and the small size of the population resulting in one of the highest lengths of public road per taxpayer in the Western motorised world. Infrastructure cost is paramount. This helps explain the resistance to Safe System investment and the reluctance to move away from trade-off decision making. Matching travel speeds to extant levels of infrastructure safety is thus especially critical. A Safe System approach to speed management requires community understanding and support if survivable crash outcomes are to be achieved.3

While ongoing efforts are certainly required to achieve alert and compliant road users, major continuing benefits will be achieved through focusing on road network safety improvements (achieving forgiving infrastructure) in conjunction with reviews of posted speed limits (to be set in response to the level of protection offered by the road infrastructure) and by the progressive introduction into the fleet of modern vehicle safety features. Promising crash avoidance safety features include, for example, ESC, auto emergency braking, Intelligent Speed Assist and active braking systems.16 These enhanced vehicle safety features are becoming available at a remarkable rate. Encouragingly, at least in Australia, the public awareness of these life-saving opportunities is underpinned by extensive public education, particularly in Victoria through the monopoly government insurer, the Transport Accident Commission, demonstrating a growing market for such features.17 Major safety benefits await better informed (and perhaps incentivised) choices of used car purchases (within affordable pricing levels) being made by young drivers and their parents. The regularly updated Used Car Safety Ratings Guide is a useful practical contribution to this information challenge.18


The probability of detection, certainty of a penalty if detected and a deterrent level of penalty remain critically important. Victoria’s penalty levels (both fines and demerit points) are high by international standards, with added deterrence imposed in some cases such as speeding, where detection of 25 km/hour over the limit also leads to a 1-month loss of licence, while 35 km/hour over leads to a 6-month loss.19 Many community members have difficulty with accepting—or indeed understanding—deterrence theory. Those who have been exposed, for example, to the immediacy of reductions in drink-driving fatalities as police substantially increase RBT intensity in a region, recognise only too well the sensitivity of the driving population (and the level of related deaths and serious injury crashes) to increased general deterrence as represented by increased widespread visible enforcement.20 Enforcement is an absolute tenet of improved traffic safety—at least into the medium term until emerging vehicle technologies such as universal alcohol detectors/interlocks and mandatory speed limiters will be capable and widespread enough to require and achieve an almost universal level of behavioural compliance.3

Support for Safe System

There are a number of additional challenges for the implementation of Safe System thinking in Victoria. The infrastructure engineering fraternity must continue to change the focus of its work from building roads which accept that a certain level of death and injury will always occur, to building roads which are human-focused—which recognise human error and encourage self-correction; and if human error does occur, reduce the crash forces to survivable levels, and to levels where a road user can fully recover from the event.5 Additional funding is essential to break from the trade-off mindset that impedes progress with this change of focus.

Other challenges include addressing a Safe System for motorcyclists, aligning engineering guidelines with Safe System principles, raising awareness of the Safe System approach by key stakeholders, and considering the role of Intelligent Transport Systems, performance indicators and the use of route-based strategies (identifying performance issues based on specific routes, and planning and prioritising solutions on these routes).21 However, perhaps the greatest challenge has been to win public support for lower road travel speeds.3 5

Current approaches and solutions in Victoria

The Safe System approach seeks to consolidate existing traffic safety improvements and to generate further improvement. In doing so, it explicitly adopts an approach focused on outcomes and results,22 it forces the nature of interventions to be reconsidered, and it relies on a systematic refocusing of institutional arrangements to implement those interventions.3 Careful planning of the potential individual adjustments which could be effectively applied at specific locations, or preferably at locations or along lengths across the network, can maximise the traffic safety performance benefits achieved. These shifts in thinking, however, require and depend on a major increase in public and practitioner understanding in order to progress.3

Across Australia and New Zealand, the vast majority of open rural road speed limits on two-lane, two-way roads are 100 km/hour regardless of the inherent safety of the infrastructure. This limit applies to freeways as well as unsealed roads, with some exceptions for very curvilinear road lengths. The public have been educated in this thinking so that any change (reduction) is extremely difficult to achieve. This differs substantially from limits in Europe, the USA and Canada, where most non-freeway roads have a speed limit of 80 km/hour (or in some cases 90 km/hour). Indeed, in Australia, urban multilane arterials often have a speed limit of 80 km/hour, with signallised intersections. Many more serious injury crashes occur at these intersections than if the limit was 70 km/hour or 60 km/hour, so considerable injury and fatality reduction upside exists if the will to challenge these settings can be marshalled.

While governments seem to operate mostly at the reactive level, the leading institutions in Victoria now plan and manage at the proactive level. The adoption of Safe System principles and the development of scientifically based traffic safety strategies attest to this. As the strategies are put to governments for adoption, political judgements of what is acceptable to the community are made and the strategies are adjusted accordingly. It is this normal democratic process—driven by the prevailing community perceptions about the causes of serious crashes—that prevents the advancement to a truly generative approach to traffic safety. For example, the policies of the State depend on the politicians who consider that they ‘reflect the will of the people’. On many occasions, when there is opportunity to test community views (rarely done at the political level in an open way), the community is well ahead of their political representatives in terms of supporting progressive change.3 This assumed sense of knowing what their constituents want reflects the limits around the accuracy of understanding of opinion through personal interaction and personal polling of views conducted informally by most politicians.

Governments now control discussion papers about issues including road safety by refusing to put some issues out in the public for debate. For example, a government discussion paper on Young Driver Safety and Graduated Licensing Improvement Options was circulated in Victoria in August 2005. The paper indicated the benefits of peer passenger restrictions and late-night curfews in the first year of probationary (solo) driving. The paper also included statements from the government that neither measure would be supported.23 It was only with a change in Chief Minister that the passenger restriction was in fact introduced a couple of years later. The subsequent benefits of reduced deaths and serious injuries have been as expected. Often seemingly contentious issues (in the past) when tested and with open discussion are in fact, contrary to the view of cabinet, supported by a majority of the community. This is normal political behaviour, and we cannot expect politicians to change until the community stops regarding the traffic safety problem as one of individual behaviour and starts treating it as a societal problem.3

One important step is to get the key institutions within road transport (road builders, traffic managers, vehicle makers, transport planners and others) to develop generative safety cultures internally. While Safe System thinking requires institutions such as road and traffic authorities to accept a level of accountability far greater than ever applied historically, this still falls fundamentally short of the generative safety cultures that can be found in the industrial field.3 There are signs in Australia that this process has begun, but it is by no means complete. Governments and corporations do not (generally) buy the safest vehicles in the classes they need operationally; governments permit private toll road operators to seek revenue from roadside and above-road advertising, yet bemoan distraction as an issue; most police do not enforce low-level speeding (although cameras do); there is a failure to hypothecate speed fine revenue to additional safety measures in most states and territories; and there is a lack of action on vehicle advertising promoting speed and power, among many other examples.3

In effect, those who plan, design, build and operate the road transport system have very little training in safety principles or strategies and certainly not in Safe System thinking.24 We have taken so long to progress because there was only a gradual shift from action based on experience, intuition, judgement and tradition to action based on evidence, science and technology.10 Importantly, though, Victoria lags because we have a legacy of an inherently unsafe road transport system. The resource and accountability implications of transformational change, plus the challenging nature of the new realities and subsequent impacts on the travelling public, overwhelm planners, government officials and certainly the politicians.3

The more substantial challenge is dealing not with the safety of new roads, but in improving the safety of the legacy or existing roads making up the majority of the networks, and which will take decades to be adequately retrofitted (eg, adding wire rope barriers to existing roads). In these circumstances, early unpopular action on lowering speed limits for the most unsafe roads, which will not receive infrastructure safety upgrades for, say 5 or more years, is required. Motor vehicle drivers will be, in many instances, required to drive more slowly than they might like until inherent safety of the road traffic system can be assured.

The goal of no deaths or serious injuries has not yet been accepted by the public, and education to move closer towards acceptance is a vital objective requiring a long-term strategy to ensure support for system investment and ongoing speed management. Further, injury compensation and safety enforcement for workplaces are institutionally separate from transport. The taxi, bus and truck industries, couriers, and safety for road workers are all matters for the transport agency not the workplace safety agency, and compensations for injuries incurred on the journey to/from work are similarly separated.3 This is problematic on many levels and poses a real challenge for implementing Safe System thinking.


The Safe System approach was introduced internationally with the publication of the International Transport Forum Towards Zero Report in 2008, leading to its adoption as the underpinning basis for action for the United Nations Decade of Action 2011–2020.25 While the Safe System approach has been adopted as a concept for many years in Australia, implementation lags well behind the desired goal. Safe System requires a major change in thinking, and challenges agencies to reform thinking and embrace change. Political commitment has not been consistently sustained over the past two decades and continues to be volatile.26 Public opinion does not seem to have moved past the historical ‘blame the victim’ stage, and this partly explains the political reluctance to embrace the evidence base.3–5 14 27 The cost of transformational changes to infrastructure is the second key blocker.

Performance is conditional on the starting point characteristics of the problem, the road safety management capacities and institutional arrangements, and the physical and sociopolitical environments.5 It is important to have the optimal conditions in place to achieve optimal results. For road safety, getting the institutional arrangements right is fundamental, forming the foundation of effective programmes.26 These foundations must include the knowledge and technologies to inform and address the actions, the management capabilities and resources to carry out the actions, and the political and management commitment to support the actions.28

Victoria’s ability to meet the challenges associated with Safe System implementation will be in part determined by the level of political and managerial commitment and leadership that will be required of governments to pursue ambitious road safety objectives.28 Significant gains are usually hard-won and require the help of partners and community members to make a real difference.11 Australia has an international reputation for road user behaviour change programmes and practices. However, countering that achievement is the reality that Australia has very long lengths of road per taxpayer, a high-income country condition shared perhaps only with Canada.29

As noted earlier, Australian rural road and urban arterial road speed limits for roads of average safety condition are higher than in most other high-income countries, failing to respect the safety condition of those roads. Infrastructure safety in vehicle travel conditions of varying mass and kinetic energy is often an abstract concept not readily observable to the driver or rider. While difficult, greater awareness of the crash outcome risk in these circumstances by Australian road users, and promotion of acceptance of the need to change strong attitudes by many to inappropriate travel speeds, is needed.

It is necessary to build the Safe System approach into policies, operating strategies, systems, programme guidelines, project development and professional practices. However, getting started is the biggest challenge for agencies, and this requires high-level leadership, organisation-wide inputs and modification as understanding evolves. Regular review is important, as is supporting the development of other agencies and local government. Systematic and periodic monitoring is required, along with objective benchmarking of current investment performance, making evidence-based decisions to direct future investment, identifying gaps in technical knowledge and capability, and good governance in order to achieve high-level programme goals.

Overall, the impact of the Swedish and Dutch traffic safety visions on other countries has been profound. As economist Daron Acemoglu and political scientist James A Robinson pointed out in their book Why Nations Fail,

‘while the escalated level of ambition (zero deaths and serious injuries) represents a radical shift within the road sector, these targets can be viewed as consistent with the safety expectations acceptable in other modes of transport (for example the aviation and rail sectors). What was initially seen as radical and unachievable has increasingly become the benchmark for acceptable road safety results.’ 22

While Victoria has made substantial road safety gains, there is still much work to be done in the full implementation and achievement of a Safe System.

What is already known on the subject

  • Safe System thinking reframes the way we view and manage safety.

What this study adds

  • Implementation is not straightforward, and reaching a shared understanding is difficult.

  • The challenges are generally social and political rather than technological.

  • There is still much work to be done in the full implementation and achievement of a Safe System in Victoria.



  • Contributors CM conceptualised the paper and drafted the content. IRJ contributed to conceptualisation and reviewed the content. EH contributed to conceptualisation and reviewed the content.

  • Competing interests None declared.

  • Provenance and peer review Commissioned; externally peer reviewed.