High Quality Data for Smarter Transportation

That’s why our project, “Data Quality Tools for Ensuring System Reliability of Traffic Information Centres,” co-funded by the Technology Agency of the Czech Republic, set out to tackle this challenge head-on. By combining advanced data governance methodologies with AI-driven algorithms and leveraging the cutting-edge Accurity data quality tool, we developed a robust framework to enhance the accuracy, reliability, and usability of traffic data. Our work proves that better data leads to better decisions—and ultimately, to smarter, safer, and more efficient transportation systems for everyone. 

Why Does Traffic Data Quality Matter? 

Imagine a highway monitoring system that provides real-time updates on traffic flow, speed limits, and congestion. Now, imagine that this system is working with flawed data—incorrect vehicle counts, missing speed records, or outdated congestion reports. The impact of poor data quality can be felt at multiple levels: 

• Traffic management centres may struggle to adjust signal timings efficiently. 

• Road planners might make investment decisions based on incorrect traffic volume estimates. 

• Drivers could receive misleading travel time estimates, leading to frustration and inefficiencies. 

To prevent these issues, we focused on designing a systematic approach to measuring traffic data quality that can identify errors, detect anomalies, and ensure consistency across multiple data sources. 

How Did We Do It? 

The key to improving traffic data quality lies in combining rigorous methodology with practical implementation. Our project followed a structured process: 

1. Analyzing Data Sources: We worked with various datasets, including data from automatic traffic counters, floating car data (FCD), truck rest areas, weather stations, and weigh-in-motion systems. These diverse data sources provided a comprehensive picture of traffic conditions of the road network in the Czech Republic. 

2. Designing Data Quality Rules: Once oriented in the delivered traffic data, we designed a set of data quality rules specific to traffic to assess key quality dimensions, such as accuracy, completeness, timeliness, consistency, validity and uniqueness. These rules help detect missing values, formatting issues, and logical inconsistencies in the data. 

3. Implementing in Accurity Quality: To implement the designed data quality rules, we used Accurity Quality, a powerful data quality management tool. This allowed us to efficiently design, implement, and automate the designed data quality rules, ensuring seamless integration with existing traffic monitoring systems. 

4. Testing on Real Traffic Data: We applied our data quality system to actual traffic data collected from the whole road network in the Czech Republic throughout 2023. By running systematic checks month by month, we gained valuable insights into how traffic data quality fluctuates over time. 

From Error Detection to Real-World Impact 

One of the most significant findings from our project was the seasonal variation in data quality. While data quality was nearly perfect at the beginning of the year, we observed a gradual decline in later months, with missing data and anomalies appearing more frequently. Some potential reasons included: 

• Technical issues at sensor locations, leading to missing data. 

• Roadworks and infrastructure changes, affecting measurement accuracy. 

• Unusual traffic patterns, causing unexpected fluctuations in speed and vehicle counts. 

By identifying these issues early, traffic authorities and infrastructure managers can take corrective action, such as recalibrating sensors, improving data collection methods, or fine-tuning traffic monitoring algorithms by labelling data quality for further data processing. 

The impact of better data quality extends far beyond technical improvements. It directly benefits: 

• City planners, who can make informed infrastructure decisions. 

• Traffic operators, who can optimize traffic flows more effectively. 

• Drivers, who get more reliable travel time estimates and route suggestions. 

What’s Next? Scaling Up for a Smarter Future 

The success of this project is just the beginning. The methodology and the whole system for measuring data quality we developed is scalable and adaptable, meaning it can be applied to: 

• Other roads, encompassing the entire road network across the Czech Republic. 

• New types of transportation data, such as smart city mobility data. 

• Larger geographic areas, expanding beyond a single motorway to a national or even international level. 

Our goal is to enable smarter, data-driven decision-making across the transportation sector, ensuring that traffic management, infrastructure planning, and mobility services are built on a solid foundation of high-quality data. 

Let’s Drive the Future Together 

At the core of our work is a simple belief: Better data leads to better transportation systems. Whether you're a transportation planner, a data scientist, or a policymaker, ensuring the quality of traffic data is essential for building a more efficient, safer, and smarter mobility ecosystem. 

Would you like to learn more about how data governance can improve traffic management? Get in touch with us and let’s drive the future of smart transportation—together!