1678 feasibility criteria. Broadly, the literature can be grouped into: (i) simplified cost-based economic models, (ii) economic models based on block valuation, (iii) mathematical optimization models (integer and mixedinteger programming), and (iv) extensions incorporating uncertainty and additional constraints such as operational and geotechnical conditions. Early approaches framed the transition mainly as an economic problem. Abdollahisharif et al. (2008) introduced BEV/NPV-based comparisons between continuing as open pit versus switching to underground mining, aligning the assessment with block-based valuation practices. In parallel, geotechnical feasibility became a binding condition through crown pillar design requirements; Bakhtavar et al. (2010), proposed formulations to estimate crown pillar thickness based on rock mass and geometric variables. More recent state-of-the-art research has consolidated the problem as one of optimization with discrete decisions, where the objective is typically to maximize NPV subject to economic, operational, and geotechnical constraints. Chung et al. (2016) address the transition using binary integer programming (BIP) to determine the optimal switching point, incorporating constraints such as pit slope angle, crown pillar stability, and operational extraction conditions. These models enable the simultaneous evaluation of openpit and underground alternatives within an integrated framework, scaling from 2D representations to 3D models with a large number of blocks, thereby approaching realistic planning scenarios. As an extension, Chung et al. (2022) move toward mixed-integer programming (MIP) with simultaneous production optimization for both methods and explicit incorporation of transition timing, including scenarios with production delays. A distinctive feature is that the model not only assumes the crown pillar’s existence but optimizes its location and size, and it adds typical underground development investments (ramps, drifts/tunnels, and accesses) together with period-based extraction constraints. Similarly, Bakhtavar, (2017) incorporate grade uncertainty through binary stochastic programming, integrating precedence constraints and extraction capacity constraints; however, such extensions typically entail high computational complexity and often leave out economic uncertainties (prices/costs) or other nontechnical dimensions In summary, existing transition models converge toward a relatively consistent set of key decision variables: (1) economic/financial variables (costs, prices, NPV/BEV, discount rate, development investment), (2) operational variables (sequencing, period capacities, precedences, blending, delays, and continuity), (3) geotechnical and design variables (slope angle, slope stability, crown pillar location/thickness/stability, rock mass parameters), and (4) geological and metallurgical variables (grade and recovery), with some explicit incorporation of technical uncertainty in recent studies. Nevertheless, even in advanced formulations, socio-environmental dimensions are rarely incorporated as quantified criteria with traceable weights within the decision framework and, when considered, they are typically treated indirectly (e.g., through regulatory constraints) or as external considerations to the optimization model; this gap motivates complementing traditional approaches with multi-criteria decision-making methods, particularly the Analytic Hierarchy Process (AHP) (Saaty, 1990), which structures heterogeneous criteria through pairwise comparisons, derives consistent weights, and produces a transparent ranking to integrate socioenvironmental variables as explicit inputs to strategic assessments, complementing (rather than replacing) the techno-economic core of existing transition models. 3. METHODOLOGY MCDM methods provide a suitable framework for addressing problems that involve multiple heterogeneous and potentially conflicting dimensions, as they enable the structuring of the decision problem, the comparison of the relative importance of criteria, and the derivation of explicit rankings to support decision-making (Mardani et al., 2015). Within these approaches, the AHP is one of the most widely used methods for criteria prioritization in complex contexts, due to its formulation based on pairwise comparisons, its ability to systematically incorporate expert judgment, and the possibility of assessing the internal consistency of the evaluations (Saaty, 1977; Saaty & Tran, 2007). In this study, AHP is used to derive relative
RkJQdWJsaXNoZXIy MTM0Mzk2