Abbott, B & Laylavi, F 2026, 'Artificial-intelligence-enabled rehabilitation liability estimation: 
developing a simplified bond calculator for improved financial assurance', in AB Fourie, G Boggs, J Heyes & M Tibbett (eds), Mine Closure 2026: Proceedings of the 19th International Conference on Mine Closure, Australian Centre for Geomechanics, Perth, pp. 1-6, https://doi.org/10.36487/ACG_repo/2615_38
(https://papers.acg.uwa.edu.au/p/2615_38_Abbott/)
Abstract: Accurate and defensible rehabilitation liability estimation is critical to mine closure planning, regulatory approvals and the operation of robust financial assurance frameworks. The existing Victorian State Government rehabilitation bond calculator can appear complex, dataintensive and difficult for operators, particularly small- and mediumsized mines and quarries, to apply consistently. This often results in highly variable cost estimates, gaps in input data, and limited ability to track liabilities over time as site conditions and operations change. These challenges can undermine the effectiveness of closure planning and complicate rehabilitation liability estimation.
This paper presents the development of a simplified rehabilitation bond calculator supported by conversational-based artificial intelligence (AI)enabled workflows designed to improve usability, transparency and longterm tracking of rehabilitation liabilities. The approach focuses on identifying the primary cost drivers of rehabilitation activities, simplifying data entry, and reducing estimation uncertainty for both operators and the regulator. Machinelearningassisted features provide assistance, flag missing or inconsistent inputs and support rapid scenario testing for rehabilitation options. This enables users to understand the sensitivity of their estimates and update liabilities more accurately as progressive rehabilitation occurs or operational footprints change.
Resources Victoria, within the Department of Energy, Environment and Climate Action, as the regulator for the mineral and extractive industries, has developed a simplified AI-enabled rehabilitation bond calculator as an additional tool to the existing Excel-based rehabilitation bond calculator. It has been designed initially to assist the calculation of rehabilitation bonds for low- to medium-risk mining and extractive operations, with future opportunities for expansion. 
The paper outlines the methodology used to develop and validate the tool, including benchmarking against existing regulatory calculators. The integration of AI provides significant improvements in consistency, estimate defensibility and user experience while maintaining alignment with regulatory expectations and costestimation principles.
The work demonstrates how modern digital tools can support clearer, more reliable rehabilitation liability forecasts and strengthen financial assurance frameworks. The simplified, AIenabled calculator offers practical benefits for industry, regulators, and consultants by promoting early planning, reducing common estimation errors and enabling proactive management of rehabilitation costs throughout the life of the operation.

Keywords: rehabilitation liability estimation, financial assurance, mine closure planning, bond calculator, artificial intelligence, progressive rehabilitation, cost estimation, regulatory compliance, digital tools, closure forecasting