Lissette Iturburu

Structural drawings are widely used in many fields, e.g., mechanical engineering, civil engineering, etc. In civil engineering, structural drawings serve as the main communication tool between architects, engineers, and builders to avoid conflicts, act as legal documentation, and provide a reference for future maintenance or evaluation needs. They are often organized using key elements such as title/subtitle blocks, scales, plan views, elevation view, sections, and detailed sections, which are annotated with standardized symbols and line types for interpretation by engineers and contractors. Despite advances in software capabilities, the task of generating a structural drawing remains labor-intensive and time-consuming for structural engineers. Here we introduce a novel generative AI-based method for generating structural drawings employing a large language model (LLM) agent. The method incorporates a retrieval-augmented generation (RAG) technique using externally-sourced facts to enhance the accuracy and reliability of the language model. This method is capable of understanding varied natural language descriptions, processing these to extract necessary information, and generating code to produce the desired structural drawing in AutoCAD. The approach developed, demonstrated and evaluated herein enables the efficient and direct conversion of a structural drawing's natural language description into an AutoCAD drawing, significantly reducing the workload compared to current working process associated with manual drawing production, facilitating the typical iterative process of engineers for expressing design ideas in a simplified way.

Biannual inspections are required to assess the physical and functional condition of our nation’s bridges. The Federal Highway Administration (FHWA) and various Departments of Transportation (DOT) in the United States periodically update specifications and techniques to normalize and advance the bridge inspection procedures. However, some ambiguity remains in these inspection requirements. One relevant example relates to inspection intervals and techniques. Currently, FHWA requires routine bridge inspection at least every two years, and if necessary, inspectors can adjust the inspection frequency. The details of how one would adjust the inspection frequency is not specified. And while many advanced techniques, e.g., ultrasonic surface wave and AI-based image inspection methods, can be applied to inspect bridges, the rationale to use these techniques relies on bridge inspectors’ experience. This study focuses on developing a reinforcement learning-based method to assist inspectors in managing bridge inspection planning. In this method, a reinforcement learning algorithm is utilized to optimize the frequency of inspection and the selection of the inspection method. A physics-based damage development model is utilized to simulate the deterioration process of the bridge. The reward function designed in the reinforcement learning process considers both economic cost and inspection plan risk. After training, the reinforcement learning agent can rapidly determine an optimal bridge inspection policy based on a bridge’s state, which can minimize both the cost and the risk of bridge inspection work. Thus, inspectors can refer to this agent to make a specific inspection plan for each bridge based on a bridge’s design, history, and features.

With the overwhelming number of older reinforced concrete buildings that need to be assessed for seismic vulnerability in a city, local governments face the question of how to assess their building inventory. By leveraging engineering drawings that are stored in a digital format, a well-established method for classification reinforced concrete buildings with respect to seismic vulnerability, and machine learning techniques, we have developed a technique to automatically extract quantitative information from the drawings to classify vulnerability. Using this technique, stakeholders will be able to rapidly classify buildings according to their seismic vulnerability and have access to information they need to prioritize a large building inventory. The approach has the potential to have significant impact on our ability to rapidly make decisions related to retrofit and improvements in our communities. In the Los Angeles County alone it is estimated that several thousand buildings of this type exist. The Hassan index is adopted here as the method for automation due to its simple application during the classification of the vulnerable reinforced concrete buildings. This paper will present the technique used for automating information extraction to compute the Hassan index for a large building inventory.

The Federal Highway Administration (FHWA) mandates biannual bridge inspections to assess the condition of all bridges in the United States. These inspections are recorded in the National Bridge Inventory (NBI) and the respective state’s databases to manage, study, and analyze the data. As FHWA specifications become more complex, inspections require more training and field time. Recently, element-level inspections were added, assigning a condition state to each minor element in the bridge. To address this new requirement, a machine-aided bridge inspection method was developed using artificial intelligence (AI) to assist inspectors. The proposed method focuses on the condition state assessment of cracking in reinforced concrete bridge deck elements. The deep learning-based workflow integrated with image classification and semantic segmentation methods is utilized to extract information from images and evaluate the condition state of cracks according to FHWA specifications. The new workflow uses a deep neural network to extract information required by the bridge inspection manual, enabling the determination of the condition state of cracks in the deck. The results of experimentation demonstrate the effectiveness of this workflow for this application. The method also balances the costs and risks associated with increasing levels of AI involvement, enabling inspectors to better manage their resources. This AI-based method can be implemented by asset owners, such as Departments of Transportation, to better serve communities.