Hasil untuk "Dynamic and structural geology"

Tianhao Qian, Zhuoxuan Li, Jinde Cao et al.

Efficient deep learning traditionally relies on static heuristics like weight magnitude or activation awareness (e.g., Wanda, RIA). While successful in unstructured settings, we observe a critical limitation when applying these metrics to the structural pruning of deep vision networks. These contemporary metrics suffer from a magnitude bias, failing to preserve critical functional pathways. To overcome this, we propose a decoupled kinetic paradigm inspired by Alternating Gradient Flow (AGF), utilizing an absolute feature-space Taylor expansion to accurately capture the network's structural "kinetic utility". First, we uncover a topological phase transition at extreme sparsity, where AGF successfully preserves baseline functionality and exhibits topological implicit regularization, avoiding the collapse seen in models trained from scratch. Second, transitioning to architectures without strict structural priors, we reveal a phenomenon of Sparsity Bottleneck in Vision Transformers (ViTs). Through a gradient-magnitude decoupling analysis, we discover that dynamic signals suffer from signal compression in converged models, rendering them suboptimal for real-time routing. Finally, driven by these empirical constraints, we design a hybrid routing framework that decouples AGF-guided offline structural search from online execution via zero-cost physical priors. We validate our paradigm on large-scale benchmarks: under a 75% compression stress test on ImageNet-1K, AGF effectively avoids the structural collapse where traditional metrics aggressively fall below random sampling. Furthermore, when systematically deployed for dynamic inference on ImageNet-100, our hybrid approach achieves Pareto-optimal efficiency. It reduces the usage of the heavy expert by approximately 50% (achieving an estimated overall cost of 0.92$\times$) without sacrificing the full-model accuracy.

Jonatas Sosnoski, Gracieli Dienstmann, Helena Paula Nierwinski et al.

Limitations of the cone penetration test, especially to accurately determine undrained shear strength (<i>S_u</i>) in soft soil deposits with high in situ stresses, have motivated the development of alternative devices, such as the T-bar and ball penetration tests, commonly referred to as flow penetrometers. These devices can estimate, in a single test, both the undrained shear strength (<i>S_u</i>) and the remolded strength (<i>S_ur</i>). When equipped with pore pressure sensors, they also provide valuable information on soil stratigraphy and consolidation parameters, making them versatile tools for characterizing soft soils. This study presents the development of two flow penetrometers, piezoball and piezo-T, highlighting relevant aspects of their design and calibration, followed by experimental campaigns conducted in two Brazilian clay deposits (Tubarão/SC and Sarapuí/RJ). Field tests enabled a direct comparison between the flow penetrometers and conventional methods, both in terms of <i>S_u</i> and <i>S_ur</i>. The investigation also examined the coefficient of consolidation of the soft soils. The results demonstrate good repeatability and consistent values for the bearing capacity factors (<i>N_b</i> and <i>N_t</i>) and remolded behavior (<i>N_b-rem</i> and <i>N_t-rem</i>). Regarding the performance of the pore pressure transducers, the piezoball test demonstrated good performance in pore pressure measurements and derived coefficients of consolidation. In contrast, despite the proposed design modifications, the piezo-T exhibited instability in the readings. Although the findings are derived from specific sites, the discussion is framed in light of the ranges reported internationally, highlighting potential local implications and reinforcing the need to expand robust geotechnical databases to support future applications.

Nikolaos Kolezakis, Stefano Aime, Raffaele Pastore et al.

We investigate the rheology, microscopic structure, and dynamics of an industrially relevant dispersion made of cationic surfactant vesicles, from dilute to concentrated conditions. We find that these suspensions exhibit a shear-thinning behavior at relatively low concentrations. At the microscale, this corresponds to a well-defined transition in both the structure, marked by the appearance of a peak in the static structure factor, and the dynamics, which slow down and develop a two-step decay in the correlation functions. This low-concentration transition is particularly surprising in light of experiments showing that for surfactant vesicles of similar composition the interactions should be purely repulsive. This leads us to propose that the observed structural and dynamic transition could arise, as an entropic effect, from the large sample polydispersity coupled to crowding. The shear-thinning behavior is thus interpreted as the nonlinear response of this transient structure to the imposed flow. Our work suggests that similar effects might be a generic feature of dense, highly polydisperse charged suspensions.

Yu Lei, Peng Wang

Electron matter interactions in electron microscopy produce both elastic and inelastic scattering, forming the basis for imaging and spectroscopy. However, the integration of electron energy loss spectroscopy (EELS) with 4D-STEM and electron ptychography remains challenging because of detector geometry conflicts. Song et al. solved this issue by introducing a hollow type pixelated detector that enables hollow ptychography and allows low angle electrons to go through to the EELS spectrometer. The single-slice approach of hollow ptychography proves effective for 2D thin materials but struggles with multiple scattering in thicker specimens. Here, we introduce multislice hollow ptychography (MHP), a robust imaging modality that overcomes these limitations by accounting for multiple scattering. MHP enables high-resolution structural imaging from hollow diffraction patterns while remaining compatible with simultaneous EELS acquisition. It potentially can provide sub-angstrom lateral resolution at intermediate doses and supports full 3D atomic-layer reconstruction at ultrahigh doses, with up to 70% of total electrons available for spectroscopy. This flexible framework facilitates correlative 3D imaging and chemical mapping in complex materials, including interfaces, defects, and dopants.

Bertuel Tangue Ndawa

We consider a bi-Lagrangian structure $(ω,\mathcal{F}_{1},\mathcal{F}_{2})$ on a manifold $M$, that is, $(M,ω,\mathcal{F}_{1},\mathcal{F}_{2})$ is a bi-Lagrangian manifold. We prolong bi-Lagrangian structures on $M$, and lift a dynamic on its tangent and cotangent bundles in different ways. In some cases, we show that the lifted structures are affine. In the case of the 2-dimensional torus, we find that an extension of the same dynamic on pairs of so-called Cherry vector fields induces a conjugation action on a subset of Cherry maps (circle maps with a flat). Additionally, we define the linear connections for certain Cherry maps.

Yinzheng Zhao, Zhihao Zhao, Rundong Jiang et al.

Purpose: To introduce novel dynamic structural parameters and evaluate their integration within a multimodal deep learning (DL) framework for predicting postoperative visual recovery in idiopathic full-thickness macular hole (iFTMH) patients. Methods: We utilized a publicly available longitudinal OCT dataset at five stages (preoperative, 2 weeks, 3 months, 6 months, and 12 months). A stage specific segmentation model delineated related structures, and an automated pipeline extracted quantitative, composite, qualitative, and dynamic features. Binary logistic regression models, constructed with and without dynamic parameters, assessed their incremental predictive value for best-corrected visual acuity (BCVA). A multimodal DL model combining clinical variables, OCT-derived features, and raw OCT images was developed and benchmarked against regression models. Results: The segmentation model achieved high accuracy across all timepoints (mean Dice > 0.89). Univariate and multivariate analyses identified base diameter, ellipsoid zone integrity, and macular hole area as significant BCVA predictors (P < 0.05). Incorporating dynamic recovery rates consistently improved logistic regression AUC, especially at the 3-month follow-up. The multimodal DL model outperformed logistic regression, yielding higher AUCs and overall accuracy at each stage. The difference is as high as 0.12, demonstrating the complementary value of raw image volume and dynamic parameters. Conclusions: Integrating dynamic parameters into the multimodal DL model significantly enhances the accuracy of predictions. This fully automated process therefore represents a promising clinical decision support tool for personalized postoperative management in macular hole surgery.

Natalie Bursztyn, Ashley D'Antonio, Bart Masters

The transition to online learning during the COVID-19 pandemic highlighted the need for more accessible and inclusive alternatives to traditional field-based geoscience education. In 2020, a web-hosted desktop virtual reality interface to take strike and dip (S&D) measurements was piloted with generally positive results from user feedback. However, the platform was developed to be used in an immersive virtual reality (iVR) environment. This paper presents a case study evaluating the user experience and learning outcomes of the iVR S&D program designed to facilitate teaching geological mapping concepts. Twelve students participated in the study, completing five mapping exercises within the iVR environment and providing feedback through pre- and post-mapping surveys and structured interviews focusing on the user experience and perceptions of S&D in iVR. We found that students performed well on the mapping exercises, the iVR environment enhanced their learning, and scaffolded assignments and learning community contributed to their self-efficacy. While acknowledging the limitations of iVR in replicating real-world field conditions, we find that it is successful at building student skills and confidence while demonstrating the potential of iVR S&D as an effective accessible tool for teaching geological mapping concepts and improving spatial reasoning skills.

Ce Liu, Jun Wang, Zhiqiang Cai et al.

Despite significant progress in static protein structure collection and prediction, the dynamic behavior of proteins, one of their most vital characteristics, has been largely overlooked in prior research. This oversight can be attributed to the limited availability, diversity, and heterogeneity of dynamic protein datasets. To address this gap, we propose to enhance existing prestigious static 3D protein structural databases, such as the Protein Data Bank (PDB), by integrating dynamic data and additional physical properties. Specifically, we introduce a large-scale dataset, Dynamic PDB, encompassing approximately 12.6K proteins, each subjected to all-atom molecular dynamics (MD) simulations lasting 1 microsecond to capture conformational changes. Furthermore, we provide a comprehensive suite of physical properties, including atomic velocities and forces, potential and kinetic energies of proteins, and the temperature of the simulation environment, recorded at 1 picosecond intervals throughout the simulations. For benchmarking purposes, we evaluate state-of-the-art methods on the proposed dataset for the task of trajectory prediction. To demonstrate the value of integrating richer physical properties in the study of protein dynamics and related model design, we base our approach on the SE(3) diffusion model and incorporate these physical properties into the trajectory prediction process. Preliminary results indicate that this straightforward extension of the SE(3) model yields improved accuracy, as measured by MAE and RMSD, when the proposed physical properties are taken into consideration. https://fudan-generative-vision.github.io/dynamicPDB/ .

Feng Wu, Yuelin Zhao, Li Zhu

How to accurately quantify the uncertainty of stochastic dynamical responses affected by uncertain loads and structural parameters is an important issue in structural safety and reliability analysis. In this paper, the conditional uncertainty quantification analysis for the dynamical response of stochastic structures considering the measurement data with random error is studied in depth. A method for extracting the key measurement condition, which holds the most reference value for the uncertainty quantification of response, from the measurement data is proposed. Considering the key measurement condition and employing the principle of probability conservation and conditional probability theory, the quotient-form expressions for the conditional mean, conditional variance, and conditional probability density function of the stochastic structural dynamical response are derived and are referred to as the key conditional quotients (KCQ). A numerical method combining the non-equal weighted generalized Monte Carlo method, Dirac function smoothing technique, and online-offline coupled computational strategy is developed for calculating KCQs. Three linear/nonlinear stochastic dynamical examples are used to verify that the proposed KCQ method can efficiently and accurately quantify the uncertainty of the structural response considering measurement conditions. The examples also compare the traditional non-conditional uncertainty quantification results with the conditional uncertainty quantification results given by KCQs, indicating that considering measurement conditions can significantly reduce the uncertainty of the stochastic dynamical responses, providing a more refined statistical basis for structural safety and reliability analysis.

Ke Cheng, Linzhi Peng, Junchen Ye et al.

Structure encoding has proven to be the key feature to distinguishing links in a graph. However, Structure encoding in the temporal graph keeps changing as the graph evolves, repeatedly computing such features can be time-consuming due to the high-order subgraph construction. We develop the Co-Neighbor Encoding Schema (CNES) to address this issue. Instead of recomputing the feature by the link, CNES stores information in the memory to avoid redundant calculations. Besides, unlike the existing memory-based dynamic graph learning method that stores node hidden states, we introduce a hashtable-based memory to compress the adjacency matrix for efficient structure feature construction and updating with vector computation in parallel. Furthermore, CNES introduces a Temporal-Diverse Memory to generate long-term and short-term structure encoding for neighbors with different structural information. A dynamic graph learning framework, Co-Neighbor Encoding Network (CNE-N), is proposed using the aforementioned techniques. Extensive experiments on thirteen public datasets verify the effectiveness and efficiency of the proposed method.

Md Jobair Bin Alam, Asif Ahmed, Md Zahangir Alam

Electrical resistivity tomography (ERT) has turned out to be one of the most applied and user-friendly geophysical methods in geotechnical and geoenvironmental research. ERT is an emerging technology that is becoming popular nowadays for investigating subsurface conditions. Multiple attributes of the technology using various electrode configurations significantly reduce measurement time and are suitable for applications even in hardly accessible mountain areas. It is a noninvasive test for subsurface characterization and a very sensitive method used to determine geophysical properties, i.e., structural integrity, water content, fluid composition, etc. This paper aimed to elucidate the ERT technique’s main features and applications in geotechnical and geoenvironmental engineering through four case studies. The first case study investigated the possible flow paths and areas of moisture accumulation after leachate recirculation in a bioreactor landfill. The second case study attempted to determine the moisture variation along highway pavement. The third case study explored the slope failure investigation by ERT. The fourth case study demonstrated the efficiency of the ERT method in the landfill evapotranspiration (ET) cover to investigate moisture variation on a broader scale and performance monitoring. In all of the four cases, ERT exhibited promising performance.

Balai Chandra Das, Aznarul Islam, Shrabani Das et al.

River-road crossings are one of the basic infrastructures for rolling the wheels of development of the modern world facilitating connectivity and overcoming geographical obstacles. Their impacts on the hydro-geomorphology of the riverine landscape and the lotic habitat is a long concerning issue amongst scientists, planners and stakeholders. The construction of a new highway bridge over the Jalangi River commenced a few years ago, experiencing occasional halts and progressing at a notably sluggish pace. However, in 2022, the construction gained momentum and was completed in 2023. The study aims to evaluate the influence of the bridge and its construction procedures on the river’s channel morphology and turbidity. We gathered data on various structural aspects of the recently built highway bridge at Krishnagar spanning the Jalangi River, a dying distributary of the Padma River. This included assessments of channel geomorphology and information to water turbidity. To assess the impact of road crossing on channel geomorphology we applied tools of w/d ratio, channel efficiency and channel asymmetry (A*). At-a-station channel geomorphology of the river was drastically impacted by the structure of the bridge and its construction processes. As the river is a dying one, the impulse of road crossing felt to the downstream insignificant. The impact of road-river crossing on water turbidity appears significant at the construction site. Non-clearance of debris and soil from river beds and banks obstructed animal movement along the river banks. In like environments where robust data forms the foundation, it is recommended to pursue additional research to identify a mitigation strategy for the way forward.

Quentin Bletery, Jean-Mathieu Nocquet

In a recent publication, we showed that a stack of all GPS time series recorded before Magnitude ≥ 7.0 earthquakes suggests that large earthquakes start with a precursory phase of accelerating slow slip (Bletery and Nocquet, 2023). While no peer-reviewed comment or publication has formally contradicted this result, informal discussion has emerged on various platforms. We present here the different elements of discussion and address them through a series of tests. In particular, it has been proposed that correcting GPS time series from network common-mode noise makes the signal vanish. We confirm this result, but we show that this common-mode filtering procedure may inadvertently remove an existing tectonic signal. Moreover, the analysis of past records indicate that the likelihood that common-mode noise produces the signal we observe is well below 1 %. Additionally, we find that the signal is maximum at the location of the impending earthquakes, and for a slip direction (rake angle) close to the one of the upcoming events. The collective outcomes of these tests make very unlikely that the signal solely arises from noise. Even though the results of our tests do not irrefutably demonstrate the existence of a precursory phase of slow slip, they do support its existence. We hope that this study will motivate further work by others to provide a definite answer to the question of the tectonic origin of the observed signal and confirm or refute that large earthquakes start with a precursory phase of slow slip.

Faradhea Safira, Eko Bayu Purwasatriya, Akhmad Khahlil Gibran

This study undertakes a comprehensive analysis of the geological conditions within a specific region, with a dual objective. Firstly, it aims to detail the geophysical attributes, structural aspects, and formation processes of the region. Secondly, it strives to establish a connection between the provenance, or origin, of the sedimentary materials in the region and its geological characteristics. The primary subjects of the research are the Waturanda and Penosogan Formations. These formations were selected due to their unique geological properties that offer insights into the geological history of the region. The primary method employed in the study was petrographic analysis. This technique, involving a microscopic examination of rocks, facilitated a detailed investigation into the rock samples from both formations, highlighting their mineralogical constitution and overall textures. The study also involved a thorough examination of the original rock or parent material of the Waturanda and Penosogan Formations. This analysis provided important information about their inherent geological attributes and formation processes. One of the key findings was the identification of the tectonic environment in which these formations were developed. The study revealed that the tectonic setting was a magmatic arc, specifically a dissected arc. This significant insight into the geological conditions has profound implications for understanding the geological evolution and history of the region, thereby enriching our knowledge of Earth's dynamic geological processes.

Leonid M. Bogomolov1, Vladimir N. Sychev, Naylya A. Sycheva

An analysis has been performed of the change in stress drops over time during the period of foreshock activity of strong earthquakes for two seismically active regions with different geodynamic settings: the Northern Tien Shan and the Southern Kuril Islands. The catalogs of earthquake dynamic parameters, DP (source ones in English publications), in these regions, with a number of events, were used as initial data. The DP catalog for the Northern Tien Shan includes 183 records of source parameters of earthquakes with magnitudes of 2.6–6.0, and the catalog for the Southern Kurils – 264 records. The stress drop values throughout a general sampling were analyzed as well as that in foreshock periods of 500 days length before the strongest earthquakes. For each region 12 such meaningful events have been specified, the magnitudes were М > 5 for the Northern Tien Shan, and М ≥ 6.5 for the Southern Kurils. The median average values of stress drops during 500-day period have been determined. The temporal variations of stress drops have been compared with changes in the b-value parameter (angular coefficient of earthquake recurrence plot) in the same observation periods. The computation of b-value for the case of the Northern Tien Shan involved the catalog data of KNET seismological network (1994–2021, more than 10 000 events), and the catalog of International Seismological Center (ISC, 1964–2000) for the Southern Kurils. In both cases, b-values were determined in 500-day moving interval with one day step. The computation gave the result that the well-known effect of b-value growth before strong earthquakes manifested itself explicitly in the considered regions. It has been established that such increase in b-value is accompanied by a decrease in the averaged stress drop values. The obtained results showed that the monitoring of the stress drop values can be used to identify the non stationary stage of the seismic regime.

Leonid M. Bogomolov, Dmitry V. Kostylev, Natalya V. Kostyleva et al.

The results of experiments on electrical sounding of the near-surface layer of the Earth's crust in the fault zone, which have involved a recording of seismoacoustic and seismic noise in the close zone near the source (the primary dipole source), are represented. The experiments were carried out in 2021–2022 in the southern part of the Central Sakhalin fault with the use of the generator of electric pulses developed at IMGG FEB RAS, output electric power being up to 3 kW. The aim was to reveal seismoacoustic signatures of the medium reaction to the soundings with current pulses of 5–13 A. The generator provided significantly higher current in the dipole than its typical characteristics in the case of soundings for electrical exploration by resistance methods, as well as in the case of conventional seismic and electrical exploration. At the same time, the range of current amplitudes was much smaller in comparison with the case of a deep sounding based on application of geophysical MHD generators or other extra high-power electric pulses units. Up to now, the inverse seismoelectric effect has remained practically unexplored at currents in the “intermediate” range of ~10 A and scale lengths of the order of few hundreds of meters. The presence or absence of the medium reaction to electrical soundings was distinguished by the records of molecular-electronic sensors developed by R-sensors LLC: the CME-6111 broadband seismometer and the hydrophone, installed at a distance of about 50 m from one of the poles of the electric dipole source. An increase in the average level of seismoacoustic noise during electrical soundings was revealed, which is essentially a variety of the inverse seismoelectric effect of the second kind (excitation of elastic waves during an electric current run in a two-phase medium). Previously, no similar signature of medium reaction to the current pulses was noted in the close zone adjacent to one of the dipole electrodes. The noise level increase occurs almost without delay after the start of electrical soundings, and this is in accordance with the previously obtained results on the responses of seismic acoustic emission to powerful current pulses, which were used for a deep sounding in the Northern Tien Shan.

Saumya Saxena, Oliver Kroemer

Enabling robots to perform complex dynamic tasks such as picking up an object in one sweeping motion or pushing off a wall to quickly turn a corner is a challenging problem. The dynamic interactions implicit in these tasks are critical towards the successful execution of such tasks. Graph neural networks (GNNs) provide a principled way of learning the dynamics of interactive systems but can suffer from scaling issues as the number of interactions increases. Furthermore, the problem of using learned GNN-based models for optimal control is insufficiently explored. In this work, we present a method for efficiently learning the dynamics of interacting systems by simultaneously learning a dynamic graph structure and a stable and locally linear forward model of the system. The dynamic graph structure encodes evolving contact modes along a trajectory by making probabilistic predictions over the edges of the graph. Additionally, we introduce a temporal dependence in the learned graph structure which allows us to incorporate contact measurement updates during execution thus enabling more accurate forward predictions. The learned stable and locally linear dynamics enable the use of optimal control algorithms such as iLQR for long-horizon planning and control for complex interactive tasks. Through experiments in simulation and in the real world, we evaluate the performance of our method by using the learned interaction dynamics for control and demonstrate generalization to more objects and interactions not seen during training. We introduce a control scheme that takes advantage of contact measurement updates and hence is robust to prediction inaccuracies during execution.

C. Lin, C. Lin, E. Kjellström et al.

<p>This study investigates present and future European heat wave magnitudes, represented by the Heat Wave Magnitude Index-daily (HWMId), for regional climate models (RCMs) and the driving global climate models (GCMs) over Europe. A subset of the large EURO-CORDEX ensemble is employed to study sources of uncertainties related to the choice of GCMs, RCMs, and their combinations.</p> <p>We initially compare the evaluation runs of the RCMs driven by ERA-interim reanalysis to E-OBS (observation-based estimates), finding that the RCMs can capture most of the observed spatial and temporal features of HWMId. With their higher resolution compared to GCMs, RCMs can reveal spatial features of HWMId associated with small-scale processes (e.g., orographic effects); moreover, RCMs represent large-scale features of HWMId satisfactorily (e.g., by reproducing the general pattern revealed by E-OBS with high values at western coastal regions and low values at the eastern part). Our results indicate a clear added value of the RCMs compared to the driving GCMs. Forced with the emission scenario RCP8.5, all the GCM and RCM simulations consistently project a rise in HWMId at an exponential rate. However, the climate change signals projected by the GCMs are generally attenuated when downscaled by the RCMs, with the spatial pattern also altered.</p> <p>The uncertainty in a simulated future change of heat wave magnitudes following global warming can be attributed almost equally to the difference in model physics (as represented by different RCMs) and to the driving data associated with different GCMs. Regarding the uncertainty associated with RCM choice, a major factor is the different representation of the orographic effects. No consistent spatial pattern in the ensemble spread associated with different GCMs is observed between the RCMs, suggesting GCM uncertainties are transformed by RCMs in a complex manner due to the nonlinear nature of model dynamics and physics.</p> <p>In summary, our results support the use of dynamical downscaling for deriving regional climate realization regarding heat wave magnitudes.</p>

A. Fedorov, V. Makrygina, Anatoly M. Mazukabzov et al.

The evaluation (according to structural and geochemical rock properties ) of the quartzites from the East Sayan quartzite-bearing area as a potential source of quartz raw material for crystalline silicon and optical glass manufacturing can significantly expand the forecast resources of this type of raw materials. The geological structure of the Irkut Formation, productive of high-purity quartzites is specified within the Oka-Urik, Urengenur and Urdagargan quartz-bearing areas; geological, mineralogical-petrographic and geochemical characteristics of the main quartzite types are given, the main morphological features of productive high-purity quartzite bodies are specified to predict their occurrence at depth. The major factors in the formation of high-purity quartzite bodies include: 1) quartzites are accumulated in the siliceous-carbonate sequence of the Middle Riphean Irkut Formation within a broad but isolated basin; 2) high-purity quartzite bodies are produced as a result of dynamic recrystallization due to the deformation of primary microquartzites resulting from the collision of the Dunzhugar island arc with the Gargan microcontinent margin. Within the western part of the East-Sayan quartz-bearing area, quartzite reserves as a potential source for silicon metallurgy and production of optical glass were estimated as 134 mln tons.

Chuyuan Cao, Kui Zhang

Cutter performance evaluation is important for shield TBM during the design and refurbishment stage. To face the challenges of choosing the proper cutters for tunneling in complex condition, according to the concept of parameter profile analysis, a systematic evaluation method was proposed. In this novel method, by comparing with their expected best value and the corresponding unacceptable limit, all of the selected performance parameters can be synthesized to assess the proximity of the overall merit of each cutter with respect to all the performances considered under all possible geological conditions. Performance indexes including cutting efficiency, structural strength, wear life, and dynamic response were individually analyzed based on linear cutting tests, finite element analyses, and theoretical calculations. Finally, a case study was carried out to demonstrate how the method is applied to find the optimal cutter among a small-scale disc cutter and two scrapers used for cutting multiple soft rock. In this case, three types of concrete specimens with distinct mechanical properties were carefully prepared to substitute for soft rock. The evaluation results show that the method has an intrinsic applicability in helping to make a reasonable trade-off between cost and cutting performance during the cutter selection process.