PHOTO: LiGe Wang, a distinguished expert in rock mechanics, showcasing his commitment to innovation and excellence in engineering.
Pioneering Insights Into Rock Dynamics And The Transformative Power Of Artificial Intelligence
LiGe Wang discusses the integration of AI in rock mechanics, addressing industry challenges and highlighting innovative solutions presented in AI for Rock Dynamics.
LiGe Wang stands at the forefront of innovation in the realm of rock mechanics and engineering, seamlessly merging advanced technology with critical scientific inquiry. His profound expertise and unwavering commitment to addressing intricate underground engineering challenges have positioned him as a leading voice in the field. As a co-author of the groundbreaking work, AI for Rock Dynamics, LiGe Wang, alongside Manchao He and Wei Yao, introduces a revolutionary perspective rooted in the power of artificial intelligence.
This seminal book, crafted as an Open Access resource, not only redefines the approach to rock dynamics but also exemplifies the potential of AI to transform academic publishing. By utilising Luffa AI—a sophisticated large language model developed in China—Wang and his co-authors have successfully integrated cutting-edge technological advances with traditional methodologies, thereby facilitating a new era of research and innovation. The comprehensive coverage of theoretical foundations, testing techniques, and practical applications encapsulates the essence of modern challenges in rock mechanics, making it an indispensable guide for professionals across civil engineering, mining, and geology.
LiGe Wang’s contributions extend beyond the pages of this illustrious publication. His role as a leader in the State Key Laboratory for Tunnel Engineering and his interdisciplinary collaborations underscore his dedication to advancing the field. Through pioneering research and a commitment to problem-driven exploration, he continues to inspire the next generation of engineers and researchers to push the boundaries of knowledge.
As we delve into this exclusive interview, we invite you to explore the insights and experiences shared by LiGe Wang, celebrating his remarkable journey and the transformative impact of AI for Rock Dynamics. Join us in recognising not only the significance of this publication but also the visionary work of an author who is shaping the future of rock mechanics through the lens of artificial intelligence.
LiGe Wang’s visionary leadership and groundbreaking research are reshaping rock mechanics, inspiring a new generation to embrace technological advancements.
Can you tell us about your journey in academia and what led you to specialise in rock mechanics?
My career centers on solving underground engineering challenges, from mining instability (soft rock deformation, rockbursts) to fault dynamics—critical barriers to China’s infrastructure growth. This urgency led me to merge rock mechanics theory with engineering geology for practical solutions.
The 110/N00 mining method revolutionized safety and efficiency, curbing resource waste and ground collapse. Dubbed China’s “third mining revolution,” it emerged from addressing flaws in traditional approaches. Furthermore, I developed the NPR (HE-bolt), which dynamically elongates under stress, transforming tunnel support in seismic zones. This innovation reflects my belief: tools must adapt to nature’s forces.
Leading the State Key Laboratory for Tunnel Engineering and collaborating globally (e.g., ISRM), I prioritized rockburst prediction and deep-earth challenges. The Key Motivations and Philosophy are:
Problem-Driven Research: My focus has always been on “pain points” in engineering, from controlling soft rock deformation to mitigating seismic risks.
Interdisciplinary Collaboration: Progress hinges on merging mechanics, material science, and on-site innovation.
Global Impact and Openess: Technologies like the HE-bolt and 110/N00 method are now benchmarks, proving that local solutions can have universal value.
This journey, fueled by curiosity and responsibility, continues as we tackle new frontiers in deep-earth exploration and climate-resilient infrastructure.
“Problem-Driven Research: My focus has always been on ‘pain points’ in engineering, from controlling soft rock deformation to mitigating seismic risks.” – LiGe Wang
What motivated you to co-author AI for Rock Dynamics and how did the idea for the book come to fruition?
The motivation stemmed from two converging trends: the exponential growth of AI applications in scientific research and the urgent need to modernize rock mechanics methodologies. Traditional approaches often struggle with the complexity of dynamic systems, such as predicting rockbursts or optimizing excavation designs. By integrating AI, we saw an opportunity to revolutionize data analysis, predictive modeling, and decision-making processes. The idea crystallized during discussions with Springer Nature and China National Publications Import & Export Corporation in 2024, who proposed leveraging their AI Pen Initiative to combine domain expertise with large language models like Luffa AI. This collaboration aimed to accelerate knowledge synthesis while maintaining rigorous academic standards.
How does AI for Rock Dynamics utilise Luffa AI, and in what ways does this technology enhance the process of publishing?
Luffa AI was instrumental in automating literature review synthesis, generating draft sections based on structured outlines, and optimizing cross-referencing across 900+ sources. Its ability to process 13,000+ images and generate coherent technical descriptions reduced manual labor by ~60%. Crucially, Luffa AI addressed challenges like hallucination mitigation through iterative human-AI feedback loops, where domain experts validated outputs and refined prompts. For publishing, this technology streamlined peer review by auto-flagging inconsistencies and suggesting contextual citations, effectively compressing a decade-long process into four months. Springer Nature’s open-access platform further amplified accessibility, aligning with global trends in AI-driven academic dissemination.
“AI acted as a ‘conceptual integrator’—for instance, natural language processing helped translate geological terminologies into engineering parameters.” – LiGe Wang
What do you consider to be the most pressing challenges in rock mechanics today, and how does your book seek to address them?
The field grapples with three key challenges:
(1) scaling laboratory findings to field-scale applications,
(2) real-time monitoring of dynamic rock behavior under extreme conditions
(3) integrating multi-physics data (e.g., thermal, hydraulic, mechanical).
AI for Rock Dynamics tackles these by presenting AI-driven solutions like hybrid physics-informed neural networks (PINNs) for extrapolating small-scale data, autonomous sensor networks for underground monitoring, and federated learning frameworks to harmonize disparate datasets. Case studies on deep geothermal systems and CO₂ sequestration illustrate how these tools enhance predictive accuracy and operational safety.
Could you share insights into the interdisciplinary collaboration involved in writing this book and the role of AI in integrating fields such as civil engineering and geology?
The project united 25 scholars from computational mechanics, AI, geology, and energy engineering. AI acted as a “conceptual integrator”—for instance, natural language processing helped translate geological terminologies into engineering parameters, while computer vision algorithms correlated seismic imaging data with fracture models. A standout example is Chapter 7, where Luffa AI synthesized inputs from geologists’ field notes and engineers’ simulation data to propose novel criteria for tunnel support design. This cross-pollination was facilitated by modular AI frameworks that mapped interdisciplinary dependencies, ensuring consistency across technical domains.
How do you envision AI transforming the future of research and innovation in rock mechanics and related disciplines?
AI will catalyze three paradigm shifts:
(1) Democratization—cloud-based AI tools will empower smaller research teams to conduct high-fidelity simulations previously restricted to supercomputers;
(2) Proactive Risk Management—reinforcement learning systems will predict and mitigate geohazards in real time, as demonstrated in our book’s case study on landslide early warning;
(3) Sustainable Innovation—generative AI will accelerate the design of eco-friendly mining techniques by optimizing resource utilization and energy efficiency. These advancements align with global priorities like the UN Sustainable Development Goals.
What was it like working with major publishing platforms such as Springer Nature, and how did this influence the development and global reach of the book?
Collaborating with Springer Nature provided access to cutting-edge digital publishing tools, including AI-enhanced metadata optimization for searchability and interactive 3D models embedded in the e-book version. Their global distribution network ensured immediate availability in 190+ countries through partnerships with Amazon, Waterstones, and academic libraries. Crucially, Springer’s emphasis on open science aligned with our goal to make rock dynamics knowledge universally accessible, particularly for developing nations grappling with infrastructure challenges. The publisher’s analytics dashboard also offered real-time insights into reader engagement, informing subsequent revisions.
What cutting-edge research or breakthroughs in rock dynamics are presented in AI for Rock Dynamics that professionals in the field should know about?
The book highlights four breakthroughs: (1) AI-Enhanced Acoustic Emission Analysis—deep learning algorithms that decode microseismic signals to predict rockbursts with 92% accuracy;
(2) Digital Twin Platforms—real-time virtual replicas of underground reservoirs for simulating fluid-rock interactions;
(3) Autonomous Drilling Systems—AI-guided rigs that adapt to heterogeneous rock strata, reducing equipment wear by 40%;
(4) Ethical AI Frameworks—methodologies to address biases in training data from historically underrepresented geological regions. These innovations are contextualized with industrial applications, from shale gas extraction to nuclear waste disposal.
As an academic mentor, how do you encourage students and early-career researchers to explore the convergence of AI and traditional engineering disciplines like rock mechanics?
I emphasize three strategies:
(1) Foundational Hybridization—mandate dual coursework in mechanics and machine learning;
(2) Challenge-Driven Hackathons—host competitions where teams use AI to solve real-world problems like predicting slope stability from limited data;
(3) Industry-Academia Sandboxes—partner with firms to provide access to proprietary datasets for AI model training.
For instance, one student project used Luffa AI to optimize blast patterns in a marble quarry, achieving a 25% reduction in material waste. Mentorship also involves demystifying AI’s “black box” nature through workshops on interpretable machine learning.
What future projects or research areas are you currently focusing on, particularly in the application of AI to rock mechanics?
Our team is pioneering three initiatives: (1) Self-Healing Underground Infrastructures: AI algorithms that coordinate drones and robotic repair systems to autonomously seal fractures in salt caverns used for hydrogen storage; (2) Quantum-AI Hybrid Models: Leveraging quantum computing to simulate molecular-scale rock-fluid interactions, which could revolutionize carbon capture and storage (CCS) efficiency/ (3) AI Pen Open-Source Expansion: A global collaborative platform for modular AI tools in rock mechanics, enabling researchers to contribute algorithms for tasks like 3D discontinuity network modeling, shale anisotropy prediction, and geothermal reservoir optimization.
