RockMassClassification GeotechnicalDesign QSystem RMR GSI TunnellingEngineering slopestability

Rock Mass Classification Systems: A Global Review of Use and Dominant Approaches

*Image showing the inherent complexity of a rock mass. Source: Kadyralieva, G.A. & Kozhogulov, K.Ch & Aitkuliev, N.A.. (2023).*

Rock mass classification systems (RMCS) have been a cornerstone of geotechnical engineering practice since the mid-20th century. Designed to provide a structured way of categorizing geological formations, these systems help engineers categorize rock masses into manageable types based on measurable or observable parameters. The aim is to inform the design of tunnels, slopes, foundations, and mining excavations by offering guidance on stability, deformation potential, and required support.

Inherent complexity of a rock mass and the various description characteristics. Source: Civil Blog

Over the past several decades, more than 30 classification systems have emerged globally. This diversity reflects both the evolving needs of engineering design and the regional differences in geological conditions. In a comprehensive study involving over 1,200 survey responses from rock engineering professionals worldwide, researchers compiled a detailed overview of system usage across different applications. The findings confirm that while many systems exist, only a few are commonly used in practice.

Rock mass classification systems showing historical development and major branches from 1950 to more recent years. Source: Cosar, 2004

Among the wide range of RMCS available, three systems have consistently emerged as the most commonly applied: the Q-system, Rock Mass Rating (RMR), and the Geological Strength Index (GSI). Each system serves a unique role in geotechnical analysis and offers practical advantages based on its structure and intended application.

The Q-System was introduced in the 1970s by Barton et al. and is widely used in tunnelling design. It relies on six input parameters including rock quality designation (RQD), joint set number, joint roughness, joint alteration, water conditions, and stress reduction factor. By combining these values into a quantitative index, the Q-system provides a reliable measure of the rock mass’s overall quality. It is particularly well-suited for determining support requirements in underground excavations.
Rock Mass Rating (RMR), developed by Bieniawski in 1973, also uses six parameters, including RQD, joint spacing, joint condition, groundwater conditions, and orientation of discontinuities. RMR provides a numerical rating of the rock mass that can be correlated to suitable support types and excavation methods. The system is used in a variety of applications, with applications ranging from tunnels to foundations and slopes.
The Geological Strength Index (GSI) was introduced by Hoek and Brown in 1997 as a more qualitative approach. Unlike the Q-system and RMR, GSI is primarily based on visual assessments of rock structure and surface conditions. It serves as an input for the Hoek-Brown failure criterion and is widely used in both slope stability analysis and underground design. GSI is designed to be adaptable and suitable for situations where detailed measurements are not available.

In addition to these three systems, several modern or regional alternatives have gained traction in specific countries. For instance, the Anisotropic Basic Quality (A-BQ) system, widely applied in China, builds upon earlier Chinese standards and introduces anisotropy considerations. The Slope Mass Rating (SMR) and Q-slope systems are notable adaptations of RMR and Q, respectively, tailored for slope engineering tasks.

Comparison of the characteristics and limitations of various RMCS. Source: Erharter, Georg & B., Neil & Hansen, Tom & Jain, Sumit & Marcher, Thomas. (2024).

The global survey revealed two dominant patterns in RMCS usage. In many countries, such as Norway, the United Kingdom, and China, a single system accounts for more than half of all reported usage. In contrast, countries like India, Canada, and Mexico demonstrate a more diverse distribution, with engineers selecting systems based on project-specific requirements or regional preferences.

Despite the maturity of existing systems, most RMCS still rely on subjective visual assessments. This introduces variability between practitioners and limits the reproducibility of results. For example, different engineers assessing the same tunnel face may assign different values based on their interpretation of discontinuities or surface conditions. This variability is considered a limitation in some applications of systems like GSI and RMR.

A notable concern is that modern sensing technologies, such as laser scanning, photogrammetry, and geophysical surveys, are not yet commonly used within RMCS workflows. Although these tools can provide valuable data about rock mass behaviour, few classification systems are designed to incorporate digital inputs directly. As a result, engineers often rely on traditional observational techniques even in technologically advanced projects.

Bar chart illustrating how many countries ranked each RMCS as their most frequently used system, highlighting the international dominance of Q-system, RMR, and GSI. Source: Erharter, Georg & B., Neil & Hansen, Tom & Jain, Sumit & Marcher, Thomas. (2024).

Looking ahead, researchers suggest that new RMCS should prioritise quantifiable inputs, transparency in data processing, and compatibility with modern design software. Future systems might also draw from machine learning techniques or large-scale databases to improve predictive accuracy. Another approach involves transitioning from fixed classification schemes and focus on continuous characterisation of rock masses, better reflecting the true complexity of geological conditions.

Sources: tunnelsandtunnelling.com, researchgate.net, tu-freiberg.de, civilblog.org