Blog post

Intelligent Tolling Systems: Key Technologies Driving the New Era of Tolling

How intelligent tolling solutions help reduce revenue leakage and improve traffic management

Updated: September 21, 2023 13 mins read Published: September 18, 2023

Toll roads are a simple concept. Users get well-maintained roads, and also for maintaining existing infrastructure.

But the sentiment around this transaction has rarely been positive: Most drivers don’t like paid roads and respond negatively when new tolls are introduced.

The complaints may be ample, but their root cause is often the same: drivers don’t feel they get enough value from paying for a premium road.

For operators, such an attitude, paired with lax toll payment evasion laws, often results in major revenue leakage in the form of ignored bills, unpaid invoices, and uncollected fares.

The traditional response of authorities and road concessionaires is hiring more staff for stricter monitoring or adding barriers for restricting road access. Yet more often than not, such measures drive only marginal improvements while further antagonizing drivers.

Is there a way to make road tolling more effective and customer-oriented? New intelligent tolling solutions attempt to do so.

How intelligent tolling systems can solve pressing industry problems

Fare collection, road condition monitoring, incident detection, and rule enforcement are among the most common challenges of toll operators.

Here’s how emerging technologies help create leaner operating models for road operators and better experiences for road users.

Minimized revenue leakage

Subpar toll collection processes, low workforce efficiency, and aging technologies stand behind pervasive revenue leakage among toll operators.

What is driving leakage?

Intelligent Tolling Systems: Key Technologies Driving the New Era of Tolling

Source: Traffic Technology Today – KPMG study shows toll operators losing revenue to inefficiencies and leakage

In particular, operators lose money due to:

  • Manual payment collection practices
  • Cash-only payment systems
  • Poor license plate recognition
  • High-maintenance tolling terminals
  • Personnel costs for manual billing
  • Lost in transit or ignored invoices

In other words, legacy toll systems paired with suboptimal operating models stand in the way of yield maximization.

Here’s a cautionary tale from the Pennsylvania road authority, who switched to All-Electronic Tolling in 2020 while laying off most of its human staff. The same year, its volume of uncollected tolls surged by 50%, reaching $155 million. Turnpike cameras couldn’t effectively detect obscured or faded license plates, resulting in failures to collect payment.

Overall, across the top 40 toll authorities by revenue in the US, over $20 billion in revenue remains owed each year, of which $2.24 billion is never collected due to paper-based invoicing and ineffective account conversion processes.

A new generation of intelligent tolling solutions relies on advanced vehicle recognition technologies powered by computer vision. Apart from scanning license plates, such systems can evaluate other vehicle features to identify the model and driver. Modern systems also support multiple contactless payment methods, ranging from card taps to in-car wallet payments powered by NFC connectivity.

With such a setup, toll road operators can collect payments on the fly to avoid both queues and the costly process of chasing non-payers.

Better user experience

Most drivers see toll payments as a necessary evil. But is there a way to increase drivers’ tolerance for tolls?

A study conducted in Spain found that the acceptability of road pricing schemes is determined by three groups of factors:

  • General socioeconomic characteristics: income, age, gender, employment status, etc.
  • Trip-related attributes: trip purpose, duration, frequency
  • General attitudes such as a driver’s perceptions and beliefs about road pricing, as well as context-specific variables

In other words: Drivers are most reluctant to pay when they see little value. And that’s the type of perception you can work with by applying various principles of behavioral economics such as framing and optimism biases.

For example, intelligent toll systems can be used to implement a price guarantee option. Drivers are more willing to pay when a toll road is less congested and offers a shorter travel time relative to alternative routes.

Toll operators can supply drivers with real-time congestion data prior to entry and offer refunds or discounts when the traffic density proves higher in reality. Such pricing models can entice more drivers to use express toll lanes, as they can be reassured by the guaranteed travel time.

An analysis by the Southeastern Transportation Research, Innovation, Development and Education Center of price guarantee strategies found that such an option can help operators secure higher revenue while maintaining desired service levels and enhancing the experience of travelers.

On the technology side, dynamic toll road pricing can be implemented with real-time traffic data processing by machine learning (ML) models.

Other dynamic pricing models made possible by intelligent tolling systems include time-of-day pricing, distance-based fees, and pricing based on a vehicle’s greenhouse gas emission levels. Speaking of which…

Better environmental outcomes

A well-optimized toll road network, especially in urban areas, can help contain pollution levels at desired targets.

New tolling solutions can help with:

  • Collecting environmental data or issuing “green” rebates. For example, the European Commission proposed giving a 50% discount on distance-based road tolls to zero-emission fleets. Germany, in turn, has already introduced a toll exemption for trucks powered by LNG (liquefied natural gas) and CNG (compressed natural gas) while imposing new CO2 emissions tariffs for trucks coming into effect on December 1, 2023.
  • Establishing and enforcing low emission zones (LEZ), which prohibit entry of high-polluting cars to specific areas. Greater London, for example, already has a 24/7 LEZ, forcing non-compliant cars to pay a daily access fee or face penalty charges. However, LEZ verifications are mostly manual, which results in occasional rule breaches. Sensor-based tolling solutions could enable efficient compliance checking.

New tolling systems come equipped with sensing technology for collecting environmental data such as noise, emission, and vibration levels. For instance, Darmstadt University recently partnered with Virtonic for a project in which researchers installed a network of fixed and mobile measuring units to capture environmental and traffic flow data. With the obtained data, the team works on creating a vision for resource-optimized transport infrastructure and sustainable mobility operations.

Fleet electrification has also increased the need for new billing solutions for en route EV charging infrastructure, such as emerging electric highways. Establishing effective payment collection for en-route charging will help road operators fund new construction projects and further promote EV adoption.

6 tech components of intelligent toll systems

So what makes a tolling system intelligent?

It’s a combination of six new technologies:

  • IoT and sensors
  • 5G wireless standard
  • V2X connectivity
  • Computer vision
  • API integrations
  • Predictive analytics

IoT and sensing technology

Sensor technology went from prohibitively expensive to largely affordable over the last decade. Road operators now have a wide range of options to choose from:

  • Traffic flow sensors
  • Inductive loop detectors
  • Magnetometer sensors
  • Magnetic detectors
  • Video detection systems
  • Microwave radar sensors
  • Passive infrared sensors
  • LIDAR sensors
  • Passive acoustic array sensors
  • Ultrasonic sensors

All of these allow for recording ample road data, ranging from traffic direction and cruising speed to road vibrations. Being internet-connected, such devices can process some of this data on the edge to make instant decisions (such as to detect speed violations) and dispatch extra intel to connected business systems (such as an intelligent transportation system).

For example, 2D Lidar sensors are low-cost, high-performance, and durable solutions for tolling gantries. Overhead sensing solutions can scan passing vehicles with high precision, automatically determining their size and weight and adjusting pricing accordingly. Such systems can also detect lane straddlers and lane changers with high accuracy, meaning you can easily impose fines on the go.

For example, Lidar hardware manufacturer Cepton and Red Fox ID co-developed a new multi-lane, free-flow tolling system called Quantum. The system can detect, track, and classify vehicles on highways under any weather conditions with 99.96% accuracy. With Quantum, road operators can bill customers in real time with free-flow, barrier-free tolling.

5G wireless connectivity

Legacy tolling systems rely on two communication standards:

  • DSRC – Dedicated Short-Range Communication (915 MHz, CEN 5.8 GHz, WAVE/G5 5.9 GHz, ISO 18000-63)
  • GNSS – Global Navigation Satellite Systems, used for tolling based on dedicated on-board units

But these aren’t ideal options. The first problem is that systems based on DSRC and GNSS technology run on special-purpose infrastructure with significant capital investments and fixed costs. The second is that these standards don’t always provide accurate vehicle position data and don’t provide any controls for ensuring toll compliance.

5G connectivity standards such as enhanced mobile broadband (eMBB) and ultra-reliable low latency communication (uRLLC) are coming as alternative solutions.

eMBB can enable throughput speeds of up to 20 Gbps, which will enable new data-driven experiences. uRLLC promises as little latency as one millisecond and high network reliability. With better connectivity, intelligent tolling solutions will be able to process more complex data to enable new road applications — from real-time toll monitoring to vehicle tracking.

5G-enabled connected vehicles also produce extra data that operators can leverage. For example, you can implement pricing based on actual emissions data, vehicle weight, or traffic conditions.

5G can also make roads safer by enabling real-time communication among vehicles and smart road infrastructure. For example, server can send automatic updates to nearby vehicles when the car suddenly brakes or alert passing vehicles when another car stops on the road. It can also share this information with a transport management system so that road operators can orchestrate a fast response, such as by updating information on driving conditions or alerting authorities about an accident.

DEKRA and Telefónica are testing how such systems can be implemented with a secure 5G connection. The two companies ran a successful pilot in Malaga, Spain, and managed to achieve a latency of milliseconds. Extra security is achieved by using a public key infrastructure (PKI) integrated into the cyber-security node of the European Commission.

V2X connectivity

Vehicle-to-everything (V2X) communication is the ultimate destination for most 5G mobility projects.

Advanced connectivity could enable faster and more secure data exchanges between connected cars and connected road infrastructure — tolling gantries, smart traffic lights, parking machines, CCTV cameras, weather monitoring systems, and digital signage, among other assets.

Over 192 million vehicles already have connectivity features. By 2027, the volume of connected cars will almost double and reach 367 million units in service.

Juniper Research

For toll road operators, V2X is the key technology to implementing gate-free road entry and dynamic compliance.

V2X removes the intermediary between a toll collection system and the passing vehicle — onboard devices, stickers, and even payment cards. Instead, toll systems talk directly to the vehicle to obtain all necessary information and process payments.

Using roadside edge devices, road operators can dynamically collect fares from passing vehicles by connecting straight to an in-car wallet.

V2X connectivity can also be used to implement dynamic road pricing models and smart zoning. For example, you can automatically identify a car entering a particular zone and send information on current parking rates. You can then automatically bill the car as it leaves by processing data on the parking duration, the vehicle’s registration status (local resident, carsharing company, etc.), and other relevant parameters. Greater connectivity also helps enforce better compliance, since you can directly bill the connected vehicle rather than hope that the driver uses a mobile parking app or the nearby parking machine (which doesn’t always happen).

V2X connectivity paired with smart road infrastructure can also help manage urban congestion better. The same edge devices can inform vehicles heading in a specific direction about current traffic conditions, available nearby parking slots, and current fares.

Intellias recently helped develop a pilot IoT-based traffic management solution powered by a cloud-based location platform and V2V, V2X, and V2I channels. The system can collect data from car sensors, connected CCTV cameras, traffic feeds, weather stations, and mobile devices. All data is processed locally in the vehicle, then aggregated and sent to the cloud. Users then receive real-time insights on road conditions, traffic accidents ahead, road obstacles, construction work, and traffic jams. This traffic management solution is now being tested with 1,000 drivers on some of the busiest roadways in Europe.

Computer vision

Road cameras, combined with edge video processing and machine learning algorithms, enable computer vision capabilities for tolling systems.

With computer vision, road operators can:

  • Dynamically collect payments
  • Identify and track violators
  • Analyze traffic flows

Compared to traditional license plate capture solutions, computer vision-based systems boast higher accuracy. Apart from scanning the license plate number, they can classify a vehicle by multiple parameters (size, class, occupancy, etc.), cross-check this information against the fare database, and collect the correct payment. Thanks to multidimensional, high-precision data capture, computer vision systems don’t require extra human interventions for image assessment, license plate verification, and billing.

For instance, Tattile recently released a compact axle recognition and vehicle counting system powered by computer vision and deep learning. The automatic number-plate recognition (ANPR) system uses two cameras per lane to ensure left- and right-sided axle analysis of vehicles cruising at up to 180 km/h (112 mph). Night and day operations are possible with an infrared (IR) illuminator.

The device performs on-device vehicle counting and classification, with all analysis done in real time. Tattile’s system is fully automatic and requires no image post-processing for billing purposes.

Intellias has also developed a similar ANPR system for urban tolling and service payments. In our case, the solution has two components — a user mobile app and a cloud platform for optical character recognition (OCR). Such a setup reduces the need for installing extra hardware other than IP cameras. The algorithms use a video feed from the IP camera (NVR, HD RAID, and OpenALPR and stream information) to recognize the passing vehicle and issue a payment request to the user’s smartphone.

Since we used the KNN algorithm instead of a neural network for number plate recognition, the system can run even on basic hardware. The accuracy, however, remains high — 94.3% on the tested data set of license plates, captured in poor weather conditions and at a long distance.

Our zero-click ANPR solution eliminates the need for cash, cards, or POS terminals at any service location en route — be it a gas station, overnight parking facility, or drive-through car wash.

Computer vision also helps implement other interesting use cases like real-time incident detection and violation management. The algorithms can be trained to detect signs of an accident or vehicle failure, then alert road managers about the incident.

For instance, our team has been working on a contactless road compliance app that complements the client’s traffic management platform. When the police arrive at the incident site, they can take a photo of the vehicle using an app and get essential data about it (plate number, color, year, model, etc.). They can also easily cross-check a driver’s claims by accessing relevant aggregated data from live cameras.

Likewise, video feed data and AI-aggregated insights can be used for real-time vehicle violation detection as well as for collecting accurate data for insurance claims.

API integrations

Application programming interfaces (APIs) enable computer systems to effectively exchange data and even some of the software’s business logic.

In simple terms: standardized APIs create interoperability between different transportation software systems.

That’s an important factor as authorities worldwide strive to implement shared standards for tolling. The EU plans to implement the European Electronic Toll Service (EETS) — a set of shared standards for operating toll roads in participating countries: Norway, Sweden, Denmark, Poland, Germany, Belgium, Hungary, Austria, Switzerland, France, Italy, Bulgaria, Spain, and Portugal.

Instead of using national toll collectors, users can opt for the services of any authorized EETS provider. The EETS provider, in turn, has to ensure accurate calculation and collection of tolls, then settle transactions between the user and the respective road authority. Open APIs will help facilitate participation in the EETS scheme.

APIs can also help consolidate tolling data in markets with high fragmentation and low interoperability among tolling providers. For example, ClearRoad created a tolling API for the US market that provides drivers with full access to toll charges and toll payments across the country through a single integration point. Using this API, fleet managers can register their vehicles with multiple agencies and set up billing accounts from one interface. Toll operators, in turn, benefit from streamlined revenue collection and a reduced risk of delinquent accounts.

Apart from establishing better integration with other market players and complying with regulations, APIs can allow road managers to integrate data from:

  • Urban transportation management systems
  • Vehicle telematics systems
  • Weather stations
  • Insurance companies

With a new stream of real-time insights, you can better manage congestion. For example, you can implement speed limits based on weather conditions and display them on digital signs to minimize accidents. Or you can model how different maintenance events will affect traffic throughput based on historical data.

In short, APIs can provide tolling agencies and road managers with a wealth of insights for real-time traffic monitoring and even traffic predictions.

Predictive analytics

Ample data, combined with ML and DL algorithms, helps road managers not only analyze the past but also glean insights into the future.

With predictive analytics, you can:

  • Estimate lane throughput under given traffic conditions
  • Analyze how dynamic pricing scenarios affect revenue
  • Evaluate various traffic management scenarios
  • Test different traffic optimization ideas before implementing them

Indra, for example, recently implemented a dynamic tolling system for Israel with intelligent transportation planning, compliance enforcement, and communication capabilities. Using a mobile app, drivers can declare if they’re traveling in a high-occupancy vehicle (HOV) to qualify for a toll exemption.

For all other vehicles, the system automatically calculates rates based on real-time traffic conditions. When roads get crowded, tolls are higher, and when traffic decreases, the rates drop too. The system also integrates with various intelligent transportation systems to provide a detailed view of road conditions. Operators can then provide dynamic alerts to drivers, direct them towards free or paid routes, or suggest taking public transport instead.

AI algorithms can also help implement a range of micro-rolling scenarios to better regulate urban congestion. For example, drivers traveling during rush hour can be charged higher fees to discourage vehicle use. Those traveling during off-peak hours, in turn, can be motivated with lower fees or special rebates. Likewise, operators can discount less crowded but longer routes to alleviate congestion on shorter but more popular roads to further regulate traffic.

Tolling transformations

As sand fills the open space in a jar of stones, new tolling technologies address decades-old needs left unfulfilled by traditional systems.

An increased state of vehicle connectivity, paired with commoditized sensing technologies and analytics solutions, helps road operators gain real-time visibility into road conditions and enforce compliance from afar.

The latest generation of tolling systems also requires less infrastructure investment. Bulky, high-cost payment terminals can be replaced with zero-click, software-based payment experiences powered by ANPR. Compact roadside edge devices and regular CCTV cameras are now sufficient to perform real-time traffic flow analysis and free-flow tolling.

What’s more, new tolling technologies enable a new breadth of billing models, which many consumers find way more satisfactory than the standard ones.

Are you working on transforming the tolling market? Intellias helps companies launch new software products on the edge of future mobility. Contact us to learn more about our engineering support.

Rate this article
5.0/5.0 Thank you for your vote. 64460 b6a98ca33b

How can we help you?

Get in touch with us. We'd love to hear from you.

    I give consent to the processing of my personal data given in the contact form above as well as receiving commercial and marketing communications under the terms and conditions of Intellias Privacy Policy.

    We use cookies to bring best personalized experience for you.
    By clicking “Accept” below, you agree to our use of cookies as described in the Cookie Policy

    Thank you for your message.
    We will get back to you shortly.