America's turnpike and tolling authorities manage a massive responsibility: operating crucial infrastructure that generates more than $20 billion annually. These toll roads, bridges, and tunnels serve millions of daily travelers, all without a single tax dollar.
But this success story faces mounting pressure: Rising maintenance costs, inflation, and expanding policy mandates converge with escalating public expectations for faster, contactless, and more reliable travel experiences.
How can agencies meet these escalating demands without proportional increases in budgets or staff? With more than 86 million electronic toll collection devices in use nationwide and major modernization projects underway in states like Ohio and Kansas, leaders must find ways to improve performance without adding more resources. The answer lies in leveraging these existing systems not just as operational tools, but as data generators. With the help of artificial intelligence (AI), agencies can transform this data into actionable insights that optimize costs, enhance sustainability, and improve operational effectiveness.
Forward-thinking tolling agencies are already using AI to unlock greater safety, efficiency, and revenue protection. In this blog, we break down three real-world case studies highlighting the results achieved and the key lessons today's tolling executives can take away.
AI as a Strategic Advantage in Tolling
Artificial intelligence transforms operational data into predictive insights, enabling leaders to improve safety, streamline operations, and protect revenue while maximizing the potential of existing resources. When properly implemented, it also becomes a strategic intelligence system that allows tolling executives to answer the questions that drive competitive advantage:
- Incident Prevention: Predicting where and when crashes will occur, allowing proactive intervention
- Capacity Optimization: Maximizing throughput from existing infrastructure through intelligent traffic management
- Revenue Protection: Identifying and closing revenue leaks without imposing additional costs on travelers
The following examples demonstrate how forward-looking authorities are turning these capabilities into measurable competitive advantages.
Leading Tolling Authority: From Reactive to Proactive Safety
A major tolling agency was spending 200 hours per cycle manually analyzing crash data, a process that took 6-12 months to identify dangerous patterns, often after multiple incidents had occurred. Working with North Highland, the agency deployed an AI-Powered predictive analytics platform that integrates six critical data sources:
- CADS (Crash Analysis and Reporting System) for historical incident data
- AccuWeather for real-time and forecasted weather conditions
- INRIX for traffic flow and congestion patterns
- Plus three additional operational data streams
- CADS, AccuWeather, and INRIX
The AI system now identifies crash hotspots before incidents occur, enabling the agency to deploy targeted patrols, implement dynamic speed adjustments, and schedule preventative maintenance based on risk predictions rather than reactive schedules.
Results:
- 50% faster analysis: Risk assessment reduced from 6-12 month to 3-6 months
- Proactive intervention: Hotspots identified and addressed before incidents occur
- Continuous intelligence: Real-time dashboard enables ongoing risk monitoring and rapid response
Leadership takeaway: Early risk detection allows agencies to deploy resources where they are needed most, helping reduce accidents and strengthening public confidence in roadway safety.
Intelligent Image Processing: Increasing Throughput Without Adding Staff
The Oklahoma Turnpike Authority (OTA) faced a mounting operational challenge: manually reviewing millions of license plate images each month to validate toll transactions.
OTA's AI-powered Manual Image Review system uses computer vision, optical character recognition (OCR), and multi-stage confidence scoring to automatically process most images. Multiple AI models cross-validate results, with only ambiguous cases sent to human reviewers. The system learns continuously from human-reviewed cases, improving accuracy even in low-light or poor-weather conditions.
Results:
- 60% reduction in manual review workload freeing staff from repetitive tasks
- 98% accuracy maintained across all processed images
- Staff redeployment to higher-complexity analysis and customer service initiatives
Leadership takeaway: Intelligent automation doesn’t just reduce costs, it strategically redeploys human expertise from routine tasks to complex problem-solving, creating both operational efficiency and employee satisfaction.
Conversational Analytics: Making Data a Shared Asset
To expand access to operational data, OTA also implemented conversational analytics using natural language processing on Google Cloud Platform. Staff in any role can query databases in plain English and receive immediate, contextual insights.
For example, a manager can ask, “Which bridges in the northwest region have deterioration ratings below five?” and receive not only the answer but also relevant trends and anomalies. This shift has accelerated planning and operational decisions across departments.
Building on this success, similar conversational analytic capabilities have also been implemented for Oklahoma Department of Transportation (ODOT), where comparable work is progressing.
Results:
- Reporting time reduced by 7% (two weeks to 30 seconds)
- Non-technical staff empowered to make data-driven decisions
Leadership takeaway: Broad access to actionable data shortens decision cycles and removes information bottlenecks.
Three Imperatives for Tolling Leaders
North Highland’s multidisciplinary team of AI specialists, data architects, and transportation experts have successfully delivered AI solutions that have transformed safety, predictive maintenance, and payment collection across multiple tolling organizations. Drawing from this collective experience, three critical imperatives emerge for leaders considering AI adoption:
- Build on strong data foundations first. AI’s performance and return on investment depends heavily on high quality and well-governed data. As Doug Krauss notes, “If you do not know where your data lives or cannot trust it, AI will only magnify the problem.” For tolling authorities, this means consolidating silos, applying quality controls, and establishing clear data ownership before AI solutions.
- Start with high-impact use case. Choose use cases that have the highest impact, leading to the fastest adoption. The fastest adoption comes from projects with clear operational and financial returns. Predictive crash analytics, automated toll image review, and revenue leakage prevention deliver measurable savings while improving public service. “Start where the return is obvious,” our team advises. “It builds confidence for wider adoption.”
- Align infrastructure decisions with strategy. “Technology decisions you make today will either accelerate your progress or limit your flexibility in the future,” Krauss warns. Decisions on whether to host AI in the cloud, on-premises, or through a hybrid model must reflect long-term scalability, security, and cost objectives rather than just immediate technical requirements.
The Road Ahead for Tolling Authorities
The evidence is clear: AI is already driving measurable improvements in safety, efficiency, and revenue protection across leading tolling authorities. The question isn’t whether to adopt AI, but how quickly organizations can implement it strategically. Leaders who act now can realize these benefits in months rather than years.
For Executive Directors/CEOs, AI offers a path to meeting rising service expectations without unsustainable costs. For CFOs, it strengthens the revenue base while controlling expenses. For CTOs, it provides a modernization platform that integrates with current systems and supports future innovation.
AI delivers role-specific value that supports organization-wide priorities:
- Executive Directors/CEOs gain a path to meeting rising service expectations without proportional cost increases
- CFOs see strengthened revenue streams and controlled operational expenses through intelligent automation
- CTOs access a modernization platform that enhances current systems while building flexibility for future innovation
In a sector where budgets are constrained and public expectations continue rising, AI has evolved from option to competitive necessity. The next wave of AI adoption in tolling will likely focus on predictive maintenance that prevents costly failures, customer experience personalization that improves satisfaction, and cross-agency data integration that creates safer, more seamless travel regional travel networks.
The transformation is already underway across the tolling industry. To learn how your authority can join the leaders realizing measurable AI benefits, connect with North Highland’s transportation and AI experts today to start your modernization journey.
