Presentation Information
[NpVBMPAo-12]AI-Based Enhancement of Traditional Condition Index Frameworks for Road Management in Argentina
*GUSTAVO MEZZELANI1, Franco Piazza1, Lucas Bresciani1 (1. ITYAC SA (Argentine))
Keywords:
AI,methodology,maintenance,condition index
Traditional road condition index frameworks used in Argentina have played a central role in pavement evaluation, maintenance planning, and budget allocation.
This paper presents an AI-based approach to enhancing these traditional methodologies through the integration of machine learning models and automated distress recognition.
In the first stage, the main objective was to modernize pavement evaluation by incorporating advanced technologies while fully preserving the existing Condition Index methodology.
This approach ensures institutional continuity and compatibility with current practices, while significantly improving data consistency, objectivity, and operational efficiency.Beyond this initial enhancement, the proposed system—called TobIAs—extends the methodology by automatically recommending the most appropriate maintenance and rehabilitation actions.
By leveraging Artificial Intelligence and embedding the expertise of experienced road engineers, TobIAs transforms traditional inspection data into actionable recommendations. In practical terms, this results in smarter data usage, improved output quality based on the original methodology, and decision-making enriched by engineering knowledge captured through AI.
The development does not stop at this first stage. A second stage rethinks pavement performance indicators to achieve a more comprehensive and reliable assessment of road conditions. New indicators are being developed using advanced survey equipment and larger datasets, incorporating not only safety and ride comfort parameters but also structural pavement performance.
This integrated approach provides a more complete representation of pavement condition and long-term behavior.
Today, the primary challenge in road management is no longer data availability. The collection, storage, and processing of large volumes of information are no longer limiting factors. The real challenge lies in transforming data into meaningful, decision-oriented knowledge.
Artificial Intelligence—such as TobIAs —is a powerful enabler of this transformation. It does not replace professional judgment, but rather enhances it by providing a robust, evidence-based foundation for transparent, consistent, and informed decision-making in road infrastructure management.
This paper presents an AI-based approach to enhancing these traditional methodologies through the integration of machine learning models and automated distress recognition.
In the first stage, the main objective was to modernize pavement evaluation by incorporating advanced technologies while fully preserving the existing Condition Index methodology.
This approach ensures institutional continuity and compatibility with current practices, while significantly improving data consistency, objectivity, and operational efficiency.Beyond this initial enhancement, the proposed system—called TobIAs—extends the methodology by automatically recommending the most appropriate maintenance and rehabilitation actions.
By leveraging Artificial Intelligence and embedding the expertise of experienced road engineers, TobIAs transforms traditional inspection data into actionable recommendations. In practical terms, this results in smarter data usage, improved output quality based on the original methodology, and decision-making enriched by engineering knowledge captured through AI.
The development does not stop at this first stage. A second stage rethinks pavement performance indicators to achieve a more comprehensive and reliable assessment of road conditions. New indicators are being developed using advanced survey equipment and larger datasets, incorporating not only safety and ride comfort parameters but also structural pavement performance.
This integrated approach provides a more complete representation of pavement condition and long-term behavior.
Today, the primary challenge in road management is no longer data availability. The collection, storage, and processing of large volumes of information are no longer limiting factors. The real challenge lies in transforming data into meaningful, decision-oriented knowledge.
Artificial Intelligence—such as TobIAs —is a powerful enabler of this transformation. It does not replace professional judgment, but rather enhances it by providing a robust, evidence-based foundation for transparent, consistent, and informed decision-making in road infrastructure management.
