The Victorian Department of Transport and Planning (DTP) is responsible for managing 24,000km of arterial roads and freeways in Victoria, forming a critical backbone of the state’s economic and social infrastructure. As
the primary asset custodian, DTP oversees the stewardship, renewal planning, and service-level performance of a network that underpins mobility, freight efficiency, and statewide resilience.
DTP adopts an enterprise asset management approach to governing its network. However, apart from pavement, bridges, major culverts and ITS assets, the department lacks comprehensive baseline inventory and
condition data across the broader asset hierarchy. Out of the 138 asset subtypes categorised under six asset classes in DTP’s Asset Service Frameworks, fewer than 20 currently hold reliable location, condition, or
performance metadata. This gap in foundational asset intelligence constrains DTP’s ability to optimise lifecycle strategies, forecast renewal demand, justify investment, and ensure evidence-based service delivery.
To address this, DTP engaged Fedasen in 2022 to capture above-ground road infrastructure asset data aligned with DTP’s asset hierarchy, attribute specifications, and lifecycle management objectives. Clear business rules
and performance thresholds were established to ensure full network coverage, data standardisation, and alignment with departmental governance requirements.
Technical Solution
Fedasen’s intelligent Road Analysor (FiRA) platform leverages advanced Artificial Intelligence and deep learning models, integrated with high-resolution aerial and 360-degree ground imagery, to automate asset detection, classification, severity scoring, and condition assessment at scale. With GIS-enabled georeferencing, FiRA produces network-wide asset intelligence, precisely mapped to each asset’s location, hierarchy, and operational context.
The platform enables DTP to access consolidated asset registers, visualise condition states, track deterioration patterns, and support scenario planning through FiRA’s web portal and geo-dashboard. By shifting from manual
inspection to intelligent automation, DTP can reallocate effort to higher-value asset lifecycle optimisation, risk mitigation, and investment prioritisation.
Data Validation Methodology
A structured, multi-layer quality assurance and control (QA/QC) framework is applied to ensure data accuracy and asset-management-grade reliability. The first validation layer is automated QA embedded within FiRA, where AI models continuously check for anomalies, completeness, and attribute consistency at the point of capture. The second layer involves expert human validation, including random sampling and attribute verification performed by Fedasen analysts and independent reviewers. This ensures asset classifications, hierarchy assignments, and measured attributes consistently achieve accuracy thresholds above 95 percent.
A comprehensive audit trail is maintained for governance and assurance, with every data point timestamped and geolocated to support traceability, compliance, and lifecycle decision support. All feedback generated
throughout the validation workflow is captured and incorporated into FiRA’s continuous improvement loop, ensuring the system matures in alignment with DTP’s evolving asset management requirements.
This methodology ensures DTP receives accurate, actionable asset intelligence that strengthens lifecycle planning, risk-based decision-making, and long-term investment optimisation