Road Safety Is Solvable - If We Stop Waiting for Crashes

Every death and serious injury on our roads changes a family, a workplace and a community. In rural places, that grief is often felt even more sharply. Long distances, higher speeds, fewer transport choices and constrained local budgets can combine to make road trauma feel inevitable.

I do not believe it is inevitable.

At Abley, we see road safety as a solvable problem. Not easy. Not simple. But solvable when we stop treating crashes as the only reliable signal of risk and start using the data, technology and engineering judgement already available to act before tragedy occurs.

That is the heart of the Safe System Approach: people will make mistakes, but those mistakes should not cost them their lives. The road network should be designed, managed and improved so that human error is anticipated and the consequences are survivable.

The old model is not enough

For decades, many safety programmes have leaned heavily on historical crash data. That work matters. When a location has a clear crash pattern, it must be addressed.

But relying on crash history alone is a poor fit for many rural networks. Crashes can be sparse, random and widely dispersed. A dangerous curve, narrow shoulder or unforgiving roadside may not show up as a black spot until after someone has been killed or seriously injured. By then, the warning sign was there all along - we just used the wrong lens to see it.

A proactive approach asks a better question: where are the conditions that make death and serious injury more likely, even if the crash has not happened yet?

What we learned in Georgia

Our recent work in Georgia is a practical example of that shift. Abley worked with The Ray, the Georgia Department of Transportation and Esri to pilot SafeRoads, calibrated for United States conditions, across rural corridors.

Rather than waiting for crash history to tell the whole story, the methodology assessed the road environment itself. It considered the features that experienced safety practitioners know matter: horizontal and vertical alignment, lane and shoulder width, roadside hazards, access density, land-use context, traffic volumes and posted speed limits.

The result was powerful because it was simple to understand. In the pilot analysis, only 6 percent of the rural network was identified as high-risk, yet that small share accounted for 20 percent of rural fatalities and serious injuries. That is exactly the kind of insight rural agencies need: a clear way to focus limited funding where it can save the most lives.

This is not abstract modelling for its own sake. It is a way to help transportation professionals see risk across the whole network, prioritise with confidence and move from diagnosis to delivery.

The solutions are practical

One reason I am optimistic is that many of the most effective rural safety interventions are neither exotic nor prohibitively expensive. They are familiar, proven treatments: better signage, curve delineation, rumble strips, high-friction surface treatments, improved shoulders, safer roadsides and clearer separation between vehicles and vulnerable road users.

The challenge is not always knowing what works. The challenge is knowing where to apply it first, how to justify the investment and how to implement across a corridor or network rather than one isolated site at a time.

That is where tools like SafeRoads, SafeCurves and SafeIntersections can change the pace of progress. SafeCurves, for example, helps identify horizontal curves where speed differentials and geometry create a heightened risk of roadway departure. For pedestrians and cyclists, land-use and transition zones can highlight places where exposure risk is growing before serious crashes establish a pattern.

With that evidence, agencies can move beyond reactive spot treatments and start delivering whole-of-network programmes that lift safety performance at scale.

A fairer path for rural and Tribal communities

Rural and Tribal communities often face a double burden. They can have some of the highest safety risks and some of the least technical capacity to assemble the data, modelling and benefit cases required for competitive funding applications.

A proactive, systemic approach helps level that playing field. It gives smaller agencies a quantitative safety story without requiring decades of perfect crash data. It helps them show why a project matters, where it sits in the wider network, what countermeasures are appropriate and what safety benefits are expected.

In Georgia, our analysis indicated the potential to save 387 lives per year across the state's rural network. Even allowing for the inevitable complexity of implementation, that kind of result should give every practitioner and policy maker confidence that the prize is real.

Better data, used well

The future of road safety will be shaped by better data - but only if that data is made accessible and actionable. Digital roadway inventories, operating speed data, standardised crash information, aerial imagery, LiDAR, street-view video and emerging AI tools all have a role to play.

Computer vision and other AI-enabled methods can help classify road and roadside assets at a speed and cost that would have been unrealistic only a few years ago. That matters because a community cannot manage what it cannot see. If we can build affordable, high-quality inventories of risk, we can help rural agencies make better decisions sooner.

But technology is not the destination. Safer roads are the destination. The value of these tools lies in turning complex data into implementable programmes: the right treatment, in the right place, at the right time.

From 'wait and see' to 'find and fix'

The biggest change we need is cultural. Too much of road safety practice still operates in a wait-and-see mode: wait for the crashes, wait for the pattern, wait for the funding case, then respond.

We should be moving much faster to a find-and-fix model. Find the risk. Quantify it. Prioritise it. Treat it. Then measure the improvement and keep going.

The road ahead

The roadmap already exists. We can identify high-risk corridors before crashes occur. We can select proven countermeasures. We can estimate safety benefits. We can help rural and Tribal communities compete for funding on a stronger evidence base. And we can implement improvements across whole networks, not just at the sites where tragedy has already happened.

That is why I remain optimistic. Road safety is not a mystery beyond our reach. It is a solvable problem when we apply the right mindset, the right data and the right delivery capability.

Road safety is solvable. The question is whether we will move quickly enough to use what we already know.

If you'd like to discuss how this applies to your network, feel free to get in touch.