Raisebore stability in deep caving: key takeaways for mine design engineers
Reviewed by Tom Sullivan

First reported on Australian Centre for Geomechanics – News
30 Second Briefing
Raisebore stability assessment for deep caving operations is shifting from standalone empirical tools to integrated workflows that combine mXrap’s Raisebore Stability App (QR and MSUS from RMD data), Rocscience UnWedge and RocTunnel3 kinematic analyses, and calibrated numerical stress–deformation modelling. The approach targets long, deep raises in highly stressed, structurally complex ground, addressing mechanisms such as face unravelling, wedge failure, spalling and time-dependent degradation along the full raise alignment. For practitioners, this means earlier screening of raise locations, more defensible diameter and alignment selection, and clearer definition of support and monitoring requirements.
Technical Brief
- mXrap Raisebore Stability App computes QR and MSUS from RMD-based logging, strength testing and stress inputs.
- Rock mass quality, structure, in situ stress, excavation geometry and time-dependent behaviour are treated as co-governing failure drivers.
- Empirical tools from McCracken & Stacey (1989) and other case histories are explicitly flagged as unreliable in deep, highly stressed, structurally complex ground.
- Structural kinematic checks use UnWedge for simplified wedge analysis and RocTunnel3 for 3D block modelling with ATV, scanline and underground mapping data.
- Numerical models track evolving major and minor principal stresses around proposed raises through multiple mining stages to assess spalling and stress-driven fracturing.
- Research method integrates rock mass classification, structural interpretation, kinematic analysis, stress evaluation and calibrated numerical modelling into a staged workflow.
- Design outputs include raise alignment and diameter selection, reaming performance expectations, support specification and targeted monitoring intervals along the raise.
Our Take
Tools like mXrap, Rocscience’s RocTunnel3 and UnWedge sit in a small subset of our 2414 research/safety-tagged pieces where numerical modelling is explicitly tied to caving and ground support design, signalling that software-driven geomechanics is becoming standard practice rather than specialist add‑on in Australian underground work.
The prominence of the Australian Centre for Geomechanics and AusIMM in Melbourne and Perth aligns with our broader Mining coverage, where Australia appears as one of the most frequent jurisdictions for deep underground case studies, giving operators there an unusually dense local knowledge base for caving risk management.
The heavy referencing of classic works (e.g. Lauffer 1988; Martin et al. 1999) alongside recent conference papers in the 11th AusIMM Underground Operators’ Conference and the Ninth International Symposium on Ground Support suggests practitioners are still calibrating modern numerical tools against empirically derived stability concepts, which can help reconcile conservative design cultures with more aggressive caving layouts.
Prepared by collating external sources, AI-assisted tools, and Geomechanics.io’s proprietary mining database, then reviewed for technical accuracy & edited by our geotechnical team.
Related Articles
Related Industries & Products
Mining
Geotechnical software solutions for mining operations including CMRR analysis, hydrogeological testing, and data management.
Tunnelling
Specialised solutions for tunnelling projects including grout mix design, hydrogeological analysis, and quality control.
CMRR-io
Streamline coal mine roof stability assessments with our cloud-based CMRR software featuring automated calculations, multi-scenario analysis, and collaborative workflows.
HYDROGEO-io
Comprehensive hydrogeological testing platform for managing, analysing, and reporting on packer tests, lugeon values, and hydraulic conductivity assessments.
GEODB-io
Centralised geotechnical data management solution for storing, accessing, and analysing all your site investigation and material testing data.