Paving the Way: How Artificial Intelligence Can Transform Predictive Maintenance

By Courtney Sawin, ITS America Policy Coordinator

State and local transportation agencies have traditionally relied on manual reports and analysis to maintain their roadways, which takes extensive time and resources away from other critical safety projects. Emerging technologies have transformed the transportation industry, making roadways safer, more efficient, and more sustainable. As new, innovative technologies are introduced and expanded on, it is crucial to integrate these technologies into existing physical infrastructure.

The capabilities of artificial intelligence (AI) have been revolutionizing the transportation industry by providing expedient traffic data analysis, closing gaps in data, and suggesting solutions for agencies to implement.  The full potential of AI in transportation, however, has yet to be realized and implemented by many local agencies. AI could be used in predictive maintenance to help save agencies time and resources, proactively address physical infrastructure issues, and save lives through detecting safety hazards.

Predictive maintenance for roadways is the practice of performing regularly scheduled maintenance to extend the lifespan of roads and prevent major issues before they arise. It often involves routine manual inspections on roadways by workers, which can be labor- and time-intensive as well as pose potential safety risks to workers. AI could streamline the process of predictive maintenance by augmenting manual inspections and alerting transportation agencies before an issue arises.

The Cost of Road Maintenance:

Over the years, the cost of highway and street maintenance has been steadily increasing. As more vehicles are on the roads, keeping up with the needs of physical infrastructure is essential in maintaining the safety of road users. In 2023 alone, the United States public sector spent approximately $137 billion on highway and roadway construction projects [1].

Maintaining roadways can often be difficult for state and local transportation agencies that lack sufficient funding and personnel to manually inspect roadways. Lowering the cost and time spent on manual inspection can help local transportation departments address road and highway maintenance issues before they become a safety hazard.

Federal, state, and local transportation agencies are not the sole entities that face financial burdens from poor roadway conditions. U.S. drivers spend an average of $3 billion a year on vehicle repairs as a result of damage from potholes [2]. Americans rely heavily on roads and highways to reach their destinations—potholes and other unsafe conditions are costly to American drivers and a proper solution is vital to lessen these burdens.

AI can help to lower the cost and time spent on big-ticket construction projects. Using AI to assist with predictive maintenance can help mitigate budget constraints faced by public transportation agencies while simultaneously providing cost saving benefits to the traveling public.

How it Works:

By incorporating AI-based technology into road condition monitoring, inspections of roads can be automated, saving state and local agencies the time and money that would be spent on sending a crew to do manual inspections, maximizing the workloads that crews can accomplish. AI-algorithms can identify wear-and-tear in pavement such as potholes, cracking, and other defects. Detecting these issues early on can enhance the longevity of pavement by alerting agencies and preventing roadway deterioration from becoming a more significant and costly problem.

Through AI, agencies can become more efficient with the allocation of their limited resources. AI can offer a comprehensive visualization of all roadways that state and local departments are responsible for maintaining and provide data analysis on roads that need or will need maintenance. Through machine learning and crowdsource imagery, artificial intelligence is able to process large volumes of data, as well as analyze images to detect damaged roadways. Once potential issue areas are detected, AI platforms can alert agencies to deploy workers to the specific segment of roadway that requires attention.

More Efficient Means More Sustainable:

Through more efficient practices and overall optimization of resources, AI use in predictive maintenance contributes to more sustainable practices in the transportation industry. Catching deteriorating road conditions before they worsen leads to minor repairs—saving energy that would have been expended on the machine heavy construction that comparatively major repairs necessitate. AI-based technologies also help to preserve the lifetime of roadways through early damage detection. Having an automated detection system enables agencies to meet the actual needs of road repairs while reducing the travel and emissions that would have resulted from sending crews to manually inspect significant amounts of roadway.  A proactive AI-based maintenance approach saves considerable energy, reducing the overall environmental impact.

A Closer Look: The Blyncsy Example:

Blyncsy is one of the leading companies that is revolutionizing the transportation industry through its artificial intelligence use. Blyncsy works with transportation agencies to monitor specific sections of freeway, residential roadway, or other areas to provide insights. Their AI models are trained using crowdsource imagery from dashcams on commercial motor vehicles [CS1] [GU2] to detect damage to roadways. Blyncsy then provides that raw data in near real-time to transportation departments. These models have the capability to automatically analyze roadways to detect potential issue areas at a frequency that cannot be reliably achieved through manual inspection alone.

In partnership with the state of Alaska’s department of transportation, Blyncsy has been able to help the state monitor over 5,000 miles of remote roadways, saving the agency both time and money. Instead of sending crews to drive 200 miles away from the nearest maintenance facility to inspect an area, Blyncsy allows the department to send them to precise locations and address issues proactively. By taking unnecessary maintenance trucks off Alaskan roadways, approximately 23,000 pounds of carbon emissions per vehicle per year are taken out of the environment. Crews are instead able to utilize the time that would have been taken driving around to do manual inspections for the actual repair and maintenance of roadways. 

Conclusion:

Implementing artificial intelligence would streamline maintenance processes and have immense benefits in the transportation sector. For transportation departments and maintenance crews, artificial intelligence would be an asset in predictive maintenance without replacing American workers. Artificial intelligence in predictive maintenance works in tandem with maintenance crews, completing maintenance inspections more efficiently and effectively, while helping state and local departments of transportation reduce their overall costs. It is crucial for states to adopt innovative technologies, such as AI, into their transportation system to improve efficiency, lower costs, and ultimately save lives.

Sources:

[1]: U.S. public construction highway and streets 2023 | Statista

[2]: Pothole Damage Costs U.S. Drivers $3 Billion Annually

[3] The information from this section comes from an interview on 8/15/24 with Blyncsy’s Marketing Director, Chris Austin. More information about Blyncsy can be found here.

---