GeoSim.AI : AI assistants for numerical simulations in geomechanics
Yared W. Bekele · 2025
Read GeoSim.AI : AI assistants for numerical simulations in geomechanics by Yared W. Bekele (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2501-14186.txt.
Introduction
Yared Bekele, writing alone, presents AI assistants for engineers who model soil and rock. The paper is from two thousand twenty-five. Such models predict how slopes and tunnels behave, but their software needs complex input files. Writing those files by hand is slow and demands deep expertise.
His suite, called GeoSim.AI, lets users give plain-language orders or images instead. The assistants turn those orders into exact scripts and input files for standard tools. They also help read the results once a run is done. Demo tasks include slope stability checks in several software packages. Bekele argues this could make expert modeling open to far more people. The suite is still growing, with more problem types planned.
Catalog record
| CREATOR | Yared W. Bekele |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 27 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2501-14186
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
GeoSim.AI : AI assistants for numerical simulations in geomechanics Yared W. Bekele Abstract The ability to accomplish tasks via natural language instructions is one of the most efficient forms of interaction between humans and technology. This efficiency has been translated into practical applications with generative AI tools now allowing users to get things done through natural language queries. The emergence of advanced Large Language Models (LLMs) marks a pivotal shift in this direction. With ongoing advancements in the field of generative AI, integrating natural language commands into sophisticated technical fields in science and engineering is becoming increasingly feasible. This paper introduces GeoSim.AI - a suite of AI assistants for numerical simulations in geomechanics - thereby demonstrating the transformative potential of generative AI in geotechnical engineering. We investigate how AI assistants powered by LLMs can streamline the process of creating complex simulation inputs and interpreting results by translating natural language instructions or image inputs into precise technical commands and scripts. This approach aims to bridge the gap between human intent and the intricate requirements of numerical modeling tools, potentially revolutionizing how researchers and engineers interact with simulation software. We present demonstrations involving AI assistants for performing slope stability analyses in various software packages. The demonstrations highlight the potential of this technology to significantly enhance productivity and accessibility in computational geomechanics. GeoSim.AI is under active development, continuously expanding the suite of AI assistants for various numerical simulation problems in geotechnical engineering. Keywords AI Assistants \cdot AI Agents \cdot LLMs \cdot Computational geomechanics \cdot Natural language instructions 1 Introduction The ability to accomplish tasks via natural language instructions is one of the most efficient forms of interaction between humans and technology. In recent years, the field of artificial intelligence (AI) has made remarkable strides, particularly in the domain of generative AI. The emergence of Large Language Models (LLMs) has revolutionized how humans interact with technology, enabling more intuitive and efficient communication through natural language. These advancements have found applications in various sectors and are continuously finding new areas of application in technical domains. The integration of AI-powered natural language interfaces with specialized technical domains represents a new frontier in human-computer interaction. This convergence has the potential to improve the usability of complex tools, methodologies and knowledge, making them more accessible to a broader range of users and applications. As AI continues to advance, the possibility of integrating natural language commands into sophisticated technical fields in science and engineering is becoming increasingly feasible. While this is an emerging field of application, the potential is demonstrated for various fields by recent research works. Zhang et al. 2024 demonstrated the application of generative AI for geospatial applications. Wiegand et al. 2024 used generative AI to build an assistant for setting up reservoir simulations. Alexiadis and Ghiassi 2024 discussed how LLMs can be used together with simulation software for physics-based modelling. Another field that stands to benefit significantly from this technological advancement is computational geomechanics. This discipline plays a crucial role in understanding and predicting the behavior of geological materials such as soils and rocks under various environmental conditions. Computational geomechanics relies heavily on numerical simulations to model complex phenomena based on software tools that often require significant time and effort to master. This presents a significant barrier to entry for students, researchers and engineers, requiring extensive domain knowledge and sometimes programming expertise to set up, run, and interpret the results effectively. In this context, the integration of generative AI assistants could significantly reduce the learning curve associated with numerical simulation software and improve productivity. This would enable researchers and engineers to focus more on the physics of the problem rather than the intricacies of simulation setup and execution, which is often different in different software packages for the same physical problem. This paper introduces GeoSim.AI, a suite of novel AI assistants that present an innovative approach to streamline and enhance numerical simulations in computational geomechanics. GeoSim.AI leverages LLM-powered AI to bridge the gap between human intent expressed in natural language and the precise technical requirements of geomechanical simulation software. As a tool, GeoSim.AI provides a user-friendly chat interface that accepts natural language and image inputs. As an approach, it employs generative AI technologies to translate these inputs into accurate simulation parameters and automate the generation of complex scripts to setup numerical simulations. One of the main aims of this paper is to explore GeoSim.AI’s potential to revolutionize the field of computational geomechanics. We seek to investigate the feasibility of using Large Language Models to interpret natural language instructions for geomechanical simulations, demonstrating how GeoSim.AI can effectively translate user queries into precise simulation inputs and analysis scripts. Through this exploration, we will evaluate both the potential benefits and challenges of integrating AI assistants into computational geomechanics workflows. Additionally, we aim to examine the broader implications of this technology for research productivity and accessibility in the field, considering how tools like GeoSim.AI could help users focus more on the practical aspect of the problem while the complex geomechanical modeling process is handled by an AI assistant. We also address cross-cutting issues such as whether this approach means adding another layer of blackbox on top of simulation software that may already be considered blackboxes for users without a detailed knowledge of how the software tools work in the background. By addressing these objectives, we aim to contribute to the ongoing dialogue about the role of AI in scientific computing and engineering, specifically focusing on its application in geomechanics. The successful implementation of such a system could not only enhance the efficiency of experienced researchers but also lower the entry barrier for newcomers to the field, potentially accelerating the pace of discovery and innovation in geomechanical research and applications. In the following sections, we will provide a background on the methodology behind GeoSim.AI’s development, present a demonstrative application of the tool, discuss our findings, and conclude with reflections on the future prospects of AI-assisted simulation workflows in geomechanics and beyond, as well as on some cross-cutting implications. 2 Architecture of GeoSimAI The GeoSim.AI system architecture represents a novel integration of artificial intelligence with traditional geomechanical simulation workflows; see Figure 1 . Central to the system is a Large Language Model (LLM), which serves as the primary interface between user intent and technical execution. The LLM is complemented by two critical knowledge repositories: a comprehensive Knowledge Base containing theoretical and practical information on geomechanics, and a Data & Tools base that houses relevant datasets and target software tools for simulations. At its core, the architecture is designed to bridge the gap between natural language instructions and complex numerical simulations in geomechanics. This section provides a detailed overview of the system’s architecture. Figure 1: Architecture of GeoSim.AI. 2.1 Natural Language Interface The Natural Language Interface of GeoSim.AI is implemented as a straightforward chat interface, providing users with a familiar and intuitive means of interaction. This interface serves as the primary point of entry for users to communicate their geomechanical simulation requirements to the system. This interface is designed to accept and interpret a wide range of geomechanical simulation requests expressed in natural language, including text, images and sketches, making the system accessible to users with varying levels of technical expertise. These inputs