AI-Enabled Framework for Portfolio-Level Asset Management Assurance in the Public Sector

Managing public infrastructure assets—schools, hospitals, transport, and social housing—is an essential government function to optimize the quality of the services from assets and extend the assets’ useful life. Portfolio-level asset management within agencies is essential to support decision-making to prioritize asset management and optimize against the financial and social objectives of the agency. Whilst the ISO 55000 series provides templates for individual asset management frameworks, there remains limited understanding of how asset management practices are operationalized across agencies and coordinated at the portfolio level. This study proposed an artificial intelligence (AI)-enabled framework for portfolio-level asset management assurance by developing a natural language processing (NLP)-driven method to understand how asset management practices are distributed and to inform decision-making by AI-assisted analysis on large-scale unstructured textual data. An empirical case study on across government portfolios within a state jurisdiction in Australia was applied. Findings suggest that the proposed framework can identify an agency’s asset management styles and portfolio-level asset management rationales through the strategic mapping. Implications for portfolio-level asset management include (1) the need to better connect operational activities with strategic objectives, (2) the deployment of AI-assisted expert elicitation evaluation tools that extend ISO 55000 series standards by identifying agency practice patterns and capability gaps through large-scale document analysis, and (3) the development of data interoperability and governance approaches for agency asset management funding requests to be better supported by evidence. This framework contributes to the corpus of government information literature by AI-assisted textual data analysis and provides innovative and analytical portfolio-level assurance services across complex government contexts.