2. STATE OF THE ART 2.1.Limitations of Traditional Analytical Approaches Analytical methods — based on nominal productivity, average utilization factors, and theoretical cycle times — present three fundamental limitations: implicit linearity, which ignores saturation and congestion effects when multiple resources share limited infrastructure; the treatment of averages as a sufficient representation, which conceals the proportion of periods with production well below plan; and the inability to capture interactions between subsystems, preventing identification of how a constraint in one propagates and amplifies across the others. 2.2.Discrete-Event Simulation in Mining Engineering DES models the production system as a sequence of discrete events that modify its state over time, incorporating stochastic variability through probability distributions fitted to historical data (Banks et al., 2005; Law, 2015). In underground mining, its applications include sizing of loading and haulage fleets (Halatchev and Knights, 2011), analysis of the impact of hang-ups in caving methods (Castro and Cuello, 2018), and evaluation of maintenance strategies in underground transport systems (Upadhyay and Askari-Nasab, 2016). The recent trend points towards more integrated models covering the complete production system, from the extraction points (drawpoints) to surface, and towards the incorporation of automation and remote operation scenarios. 2.3.The Role of Calibration in Model Credibility A simulation model that has not been calibrated and validated against real data has limited value as a decision-making tool. Calibration — the process of adjusting model parameters until its outputs are statistically equivalent to historical records — is what converts a theoretically correct model into a practical reliable tool. 3. METHODOLOGICAL FRAMEWORK 3.1.Simulation Methodology Figure 1 presents the methodological process used for the development of simulation models. The process begins with the collection and analysis of operational information, from which a conceptual model is built and validated with the client to ensure it reproduces the actual operational philosophy of the mine. On this basis, the simulation model is constructed, passing through two iterative control cycles: first the logical verification of the model and then its statistical calibration against historical data. Once the model passes both cycles, simulations and scenario analysis are executed. Results are subjected to a final validation and, if approved, lead to study conclusions and recommendations.
RkJQdWJsaXNoZXIy MTM0Mzk2