Aviation Insight Header Test


Safety Science

University researchers in Wisconsin are collaborating with transportation innovators to harness data and technology to prevent serious crashes and deaths while improving system efficiency.

By Dr. Xiao Qin, PE | Director of Institute for Physical Infrastructure & Transportation and of the Safe and Smart (S2) Traffic Lab, Lawrence E. Sivak ’71 Professor at University of Wisconsin–Milwaukee

States across the country are transforming roadways so they can be shared more safely by vehicles, bicyclists and pedestrians alike. We know that ensuring safety requires a holistic approach, which is expressed as the 4 E's – Engineering, Education, Enforcement and Emergency medical services, and more recently, a safe system approach. And approaches such as Complete Streets help create more walkable neighborhoods, more integrated transportation networks and platforms for people to move efficiently.

To enhance innovative safety benefits, the University of Wisconsin-Milwaukee Safe and Smart Traffic Lab team works with state, regional and national transportation leaders to better understand what risks exist and to develop possible solutions. Our work is grounded in the view that we can increase the impact of safety improvement efforts by blending two strategies: initiating systemic safety improvements and maintaining a sharp focus on preventing fatal or serious-injury crashes.

The safety community has increasingly realized that we cannot eliminate all crashes, but we might be able to eliminate fatal and severe injury crashes. This belief underlies the Vision Zero concept, which is being adopted in many areas of the country.

Objective: Reduce the Worst Crashes

Systemic improvements have proven highly effective in reducing severe and deadly crashes. Such measures involve installing safety-oriented infrastructure across a state’s road system broadly. Measures include, for example, adding rumble strips to the edges of the roadway or installing high-tension cable guards across many miles of highway median. Such measures help to prevent crashes that, despite being random and sporadic, tend to have severe consequences when they occur. Implementation of such measures doesn’t require a crash history; only an understanding of the factors that contribute to crashes. With strong guidelines, they can be implemented systemwide very cost effectively.

We also are exploring the effectiveness of more targeted measures that can address serious crashes. This work involves significant focus on left turns being executed across oncoming traffic, because the drivers involved often are most vulnerable to severe injuries in a crash scenario. An important method to address this risk is to channelize traffic, directing vehicles to reduce or eliminate the number of conflict points between them. This comes down to adopting innovative intersections to replace or upgrade the conventional designs we see on highways across America.

One example of this strategy is the J-turn intersection, which many states have been implementing. Instead of drivers on the minor approach needing to cross oncoming traffic to get to the opposing lane and turn left, they execute a right turn; they then travel with the traffic flow, move over to the left lane, and then make a U-turn to begin driving in their desired direction. This J-shaped path of travel reduces their risk of collision and may take less time than waiting to time their rapid acceleration across a major highway with high-speed traffic.

Another innovation, which is beginning to gain traction nationally, is the diverging diamond interchange. With this interchange design, the street traffic is directed so that it temporarily crosses to the other side of the road between highway ramps. This lets vehicles turn left onto the highway ramps without crossing oncoming lanes of traffic. As with the J-turn, this type of channelization reduces the potential conflict points between vehicles.

Roundabout Capital of USA

Wisconsin began adopting roundabouts 20 years ago as a means to improve intersection safety. It now has more roundabout intersections than any other state — about 550 of them — and is committed to building more of them in the years ahead.

The design principle of a roundabout — slow, steady traffic flow in one direction — serves to virtually eliminate fatal or serious crashes. At a conventional intersection there are 32 potential conflict points (defined by the Federal Highway Administration as any location where road users’ paths coincide) between vehicles; this drops to eight in a roundabout. There’s a similar decline in conflict points between pedestrians and motorists.

Beyond the safety dimensions to reduce serious and fatal crashes, roundabouts require less maintenance than signalized intersections, and they can make traffic flow more efficient and could reduce emissions because drivers often only have to yield, rather than stop and idle, before entering the intersection.

Data-Driven Decisions and AI Tools

The data and tools to help transportation leaders make informed decisions about safety investments continually evolve. We are entering an era when artificial intelligence applications have the potential to give transportation agencies deeper insights into safety risks and patterns, and to sharpen remedial strategies. Our research reveals some promising AI-related developments:

  • Better-Synthesized Data – The volume and dimensions of data is constantly growing. AI is able to consume high-dimension, large-volume data, and distill it so practitioners can more easily employ it in their safety planning efforts.
  • Troves of Unlocked Insights – Custom large-language models, such as Generative Pre-trained Transformer (GPT) can serve to summarize and synthesize massive volumes (millions of pages) of safety-related literature. AI thus may unlock a new universe of safety-related insights for decision-makers.
  • Near-Misses Revealed – Computer vision has been widely used in traffic surveillance systems, which lets practitioners analyze collision data after the fact. Now, by streaming a much broader range of video images and feeding other parameters into AI, transportation practitioners can gain insights into “near misses” or “close calls” too, which they can correlate to collision risk and formulate effective mitigation measures.

Moving Plans into Action

Safety is at the core of our professional practice and our commitment to the system users we serve. Here are a few thoughts that may help to guide us in our work.

1. Beyond commitment – Take concerted action

We know safety is a top priority in our industry. There can be a gap between the vision of greater safety and the actions needed to achieve it. Devote the talent and resources necessary to plan, fund, engineer and build the infrastructure that delivers real, measurable safety improvements.

2. Data management – Give it the focus it demands

Data management is a huge task. More data means more evaluation and refinement into actionable insights. This requires adopting a comprehensive data management system. It means gathering as much data as possible about crashes, roadways and traffic, and driver or telematics data, if possible. It also entails choosing the correct model to recognize patterns in the data and spot the most prominent safety risk factors among hundreds of others. Only by amassing, processing and rationalizing data can an agency pinpoint the most promising safety strategies for a specific location.

3. Countermeasures – Build a ready-to-execute library

When engineers have best practice techniques at their fingertips, they can accelerate both the progress and efficiency of safety-focused projects. Invest in creating or expanding your library of technically sound models and methods so that your team can access a range of countermeasures and relevant tools to address their specific goals.

4. Communication – Engage all stakeholders

Safety is a simple word, but manifesting it in reality is very complicated. This is why communicating the purpose, process and progress of safety initiatives is so important. Although focus is on drivers and vehicles, all users of the road, such as pedestrians and bicyclists, can contribute to crash prevention.

5. Accountability – Set goals and measure results

Every agency needs to determine what functions and individuals are accountable for achieving the safety goals that have been established. This requires distilling the broad aims of safety improvement down to certain steps, milestones and expected outcomes. With accountability comes greater commitment and results.

Partnerships for Safety

Near-term progress on reducing serious roadway crashes will rely on strong partnerships among transportation engineers, researchers, educators, public officials, technology innovators, law enforcement, insurers and many others. Our team looks forward to working with all of these participants to make Wisconsin, the U.S. and the world safer for the people who use our systems.



Dr. Xiao Qin is Director of the Institute for Physical Infrastructure & Transportation (IPIT), as well as the Lawrence E. Sivak ’71 Professor in the Civil and Environmental Engineering Department at the University of Wisconsin-Milwaukee (UWM).

Dr. Qin also is a registered Professional Engineer (PE) in Civil Engineering and the Director of the Safe and Smart (S2) Traffic Lab at UWM. He and his team at the Safe and Smart Traffic Lab explore ways to save lives, save time, and safeguard the integrity of the U.S. transportation system. Dr. Qin has received multiple research grants from Department of Transportation in past years to explore a range of safety topics. Recently, he was also given a $500,000 grant from the National Cooperative Highway Research Program to address the problem of bridge and tunnel strikes nationwide.

A thought leader in applying scientific methods to safety solutions, Dr. Qin is coauthor of Highway Safety Analytics and Modeling, a textbook which provides state-of-the-art knowledge about how to better analyze safety data, relevant tools and methods, and insights into the decision-making process for pursuing safety improvements.

He completed his Doctorate in Civil Engineering at the University of Connecticut. He earned his Bachelor of Science and Master of Science degrees in transportation and highway engineering at Southeast University in Nanjing, China.