| Abstract Scope |
A successful industrial shutdown is judged by how closely it adheres to schedule and budget. However, significant delays can be encountered when unexpected inspection results cannot be compared to legacy inspection records. In sectors like energy and manufacturing, decades of inspection and repair data are often stored in paper files or on outdated drives. Many of these documents are handwritten and not searchable, making it difficult to access crucial information when needed. As experienced personnel retire, the loss of tacit knowledge further increases the risk of delays and inefficiencies. To address this challenge, a pilot project was conducted using artificial intelligence (AI) to process thousands of legacy inspection and repair records. These documents, some dating back to the 1980s, were successfully converted into structured, searchable metadata. During a recent maintenance turnaround, this system enabled teams to retrieve key information in minutes rather than hours or days, demonstrating the potential for AI to accelerate decision-making and reduce operational risk. This example highlights how AI can support industrial workflows and preserve valuable institutional knowledge. For educators and leaders in welding and fabrication, it illustrates the importance of preparing the next generation to work with emerging technologies. This presentation will provide insight into how AI can enhance decision-making, support skilled trades, and help shape a more resilient and efficient future for the welding industry. |