Three ways municipalities benefit from new winter-weather forecasting methods

When winter weather hits, even the most experienced drivers can struggle getting from point A to point B safely. Approximately 17 percent of all vehicle crashes happen during winter conditions, and — according to the Federal Highway Administration — weather-related accidents kill 5,000 people each year on average. Consequently, for local governments, ensuring our roads are clear of snow and ice before they become impassable is crucial.

With more than 70 percent of U.S. roads seeing an average of more than five inches of snow each year, the cost to municipal governments cannot go unnoticed. Winter road maintenance forces highway agencies to spend approximately $2.3 billion each year on this work. As a result, it’s critical that agency decision-makers have access to advanced tools that help them take timely action to not only reduce road maintenance costs but, more importantly, overcome the myriad challenges winter weather poses in order to keep communities safe.

Winter Road Data Becomes a Game-Changer

Gone are the days when road maintenance decision-makers relied exclusively on TV or radio weather forecasts or field observations alerting them to deteriorating road conditions. This reactive approach puts agencies in an unfavorable position from the start because road condition reporting could quickly become out-of-date, and oftentimes, roads would become progressively more slippery and snow-covered, increasing the likelihood of accidents as a result of hazardous conditions.

Now, in order to make roads safer more quickly, decision-makers depend on weather observations, atmospheric weather models, local weather models and actual measurements from myriad sources, including radars, mobile sensors, internet of things (IoT) sensors, road weather information systems (RWIS) and environmental sensor stations (ESS). This information helps decision-makers understand road temperatures and conditions that lead to the amount of rain, snow, condensation/frost or evaporation/sublimation on the road and traffic spray, informing treatment and snow removal decisions.

However, with such a significant amount of data coming in, it has become difficult for agencies to gather all of the available information and interpret it ahead of making a decision. That’s where automation and machine learning come in to drive improvement in winter road maintenance analysis.

Advantages of Machine Learning in Analysis

With the high cost of sensors, communications and power dramatically decreasing in recent years, road maintenance decision-makers are now able to leverage automated weather analytics tools that allow for both nowcasting and forecasting to become more proactive and get ahead of the storm along city and county roads.

Today’s analytics tools employ machine learning algorithms to combine accurate real-time measurements with powerful forecast modeling capabilities and automatically display that information in an easy-to-read visual format so a decision-maker can understand how it affects their road system and use it to decide when and where to send trucks and treat roads.

The advantages of incorporating automation into winter road data analysis include the following:

1. Timely decisions, quick action. Real-time information on the type of conditions witnessed throughout the event and analysis of how pavement temperatures will act before, during and after, as well as what might happen when the snow stops falling, is critical in deciding the optimal approach for mitigating the storm’s impact. Easily accessed through web-browsers, modern analytics tools reduce the need for interpretation and simplify the decision-making process by doing the complex analysis work for decision-makers.
2. Efficient deployment of plow operators and materials. Agencies don’t want to guess whether to plow throughout the duration of an event or split the crew up to work in shifts nor do they want to pre-treat with liquids if the event begins as rain, washing away materials by the time precipitation starts to freeze and bond to the pavement. When agencies comprehensively understand the storm and its duration, making timely, targeted decisions about when, where and how many plow operators are needed and which materials will mitigate the event’s impact given the conditions becomes easier than ever.
3. Managing public expectations and comments. Social media has had a significant impact on public outreach to companies, organizations and, yes, municipalities. Questions from residents about why their street hasn’t been plowed on a very public social medium can’t be ignored like maybe a private email 15 years ago. The advent of social media has raised people’s expectations to not only complain, but to also receive relevant news updates almost immediately. Advanced analytics tools display where the trucks have been and an easy-to-read color code of current road conditions based on incoming data, which empowers decision-makers to take quick action to mitigate problem locations and keep the community better informed and therefore safer.

By analyzing and visualizing weather data, then combining relevant parameters from all sensors, automated tools provide meaningful information to help make data-driven decisions more easily. Making the right decision at the right time helps organizations optimize winter maintenance resources to save time and money, and analytics tools help managers and decision-makers at the city and county levels stay ahead of weather changes and make accurate decisions on the best approach for keeping roads safer.

Mark DeVries is a solutions manager in transportation with Vaisala. In this role, he applies his winter maintenance and chemical use expertise to help clients/agencies improve operations and solve problems.