The integration of Artificial Intelligence (AI) in engineering disciplines has brought about significant advancements, but it also necessitates the involvement of engineers to ensure safety, reliability, and productivity. This article explores specific areas where AI/ML Engineers In The Loop (EITL) are crucial across mechanical, electrical, civil, and chemical engineering among other engineering disciplines.
Mechanical Engineering
1. Design Optimization and Simulation:
Engineers play a vital role in guiding AI algorithms during the design optimization process. While AI can generate numerous design configurations, engineers must validate these designs to ensure they meet all safety and functional requirements. For instance, generative design software can produce optimized solutions, but AI/ML Engineers In The Loop need to assess these for manufacturability and structural integrity [16].
2. Predictive Maintenance:
AI systems can predict when machinery might fail, but engineers are essential for interpreting these predictions and planning maintenance activities. Subject Matter Experts in the Loop are required to validate the AI’s recommendations and ensure that maintenance actions are feasible and safe [9] [7].
3. AI-Powered Robotics:
In the deployment of AI-powered robots, Engineers are needed in the Loop to oversee the integration and operation of these systems. They ensure that robots perform tasks safely and efficiently, especially in dynamic environments where human-robot interaction is critical[9][8].
Electrical Engineering
1. Smart Grid Management:
Engineers are essential in managing AI systems that optimize power distribution in smart grids. AI/ML Engineers In The Loop validate AI predictions and decisions to ensure the stability and reliability of the power supply, addressing any anomalies that AI might miss[15][16].
2. Precision in Electronics Manufacturing:
AI can enhance precision in electronics manufacturing, but AI/ML Engineers In The Loop must oversee the process to ensure that AI-driven adjustments do not compromise product quality. They validate AI outputs and make necessary corrections based on their expertise[15][4].
3. Autonomous Systems:
For autonomous systems like drones and robots, engineers are crucial in the design, testing, and deployment phases. Engineers In The Loop ensure that AI systems operate safely and effectively, particularly in critical applications such as power line inspections[4][5].
Civil Engineering
1. Smart Construction Design:
AI tools can automate many aspects of construction design, but AI/ML Engineers In The Loop must validate these designs to ensure they comply with safety standards and regulations. They also interpret AI-generated data to make informed decisions about construction methods and materials[12][13].
2. Construction Process Orchestration:
Engineers oversee AI systems that manage construction schedules and resources. Engineers In The Loop ensure that AI recommendations are practical and align with project goals, making adjustments as necessary to address real-world constraints[2][10].
3. Safety and Risk Assessment:
AI systems can monitor construction sites for safety hazards, but engineers are needed in the AI/ML loop to interpret these alerts and take appropriate actions. They ensure that safety protocols are followed and that AI systems are correctly identifying and mitigating risks[2][13].
Chemical Engineering
1. Process Optimization:
Engineers are essential in validating AI-driven process optimizations. They ensure that AI recommendations for adjusting parameters like temperature and pressure are safe and effective, preventing potential hazards in chemical processes[1][3][6].
2. Predictive Maintenance:
In chemical plants, engineers validate AI predictions for equipment maintenance. Engineers in the AI/ML loop ensure that maintenance activities are scheduled appropriately to avoid disruptions and maintain safety standards[1][18].
3. Quality Control and Product Optimization:
Engineers oversee AI systems that monitor product quality. Subject Matter Experts are requited to validate AI-generated data and make necessary adjustments to maintain product standards, ensuring that AI systems do not overlook critical quality issues[1][6].
Conclusion
The involvement of engineers in the AI loop as subject matter experts is indispensable for ensuring the safe and productive application of AI in engineering. By combining AI’s computational power with human expertise, engineers can enhance the reliability, efficiency, and safety of engineering processes across various disciplines. This collaborative approach not only optimizes performance but also builds trust in AI systems, paving the way for more advanced and integrated engineering solutions.
For more detailed insights and examples, refer to the sources cited in this article.
Citations:
[1] https://www.dlr.de/en/ki/research-transfer/projects/ai-in-the-loop
[2] https://www.linkedin.com/pulse/role-human-intelligence-validating-ai-generated-data-nuraz-zamal-a1h0e
[3] https://cloudsecurityalliance.org/blog/2024/03/19/ai-safety-vs-ai-security-navigating-the-commonality-and-differences
[4] https://www.securitymagazine.com/articles/100378-ai-may-revolutionize-security-but-not-without-human-intuition
[5] https://snorkel.ai/human-in-the-loop-ml-fdcai-2022-daniel-wu-jp-morgan-chase/
[6] https://www.youtube.com/watch?v=ipPL6_j8MvE
[7] https://qbdgroup.com/en/blog/ai-machine-learning-validation-strategies-and-examples/
[8] https://certx.com/ai/how-human-oversight-and-transparency-can-ensure-trustworthy-ai-in-the-eu/
[9] https://ispe.org/pharmaceutical-engineering/march-april-2022/ai-maturity-model-gxp-application-foundation-ai
[10] https://www.valispace.com/ai-in-systems-engineering/
[11] https://www.siegepe.com/post/ai-in-chemical-engineering-for-natural-gas-operations
[12] https://www.monolithai.com/blog/engineering-applications-of-artificial-intelligence
[13] https://casmi.northwestern.edu/news/articles/2023/defining-safety-in-artificial-intelligence.html
[14] https://www.appliedclinicaltrialsonline.com/view/the-role-of-human-oversight-with-artificial-intelligence-in-clinical-research
[15] https://humanfactors.jmir.org/2021/2/e28236/
[16] https://www.youtube.com/watch?v=8Z7vV75Htgw
[17] https://hai.stanford.edu/news/humans-loop-design-interactive-ai-systems
[18] https://link.springer.com/article/10.1007/s44206-023-00076-w
[19] https://aiforgood.itu.int/how-ai-and-humans-cooperate-in-protecting-workers-a-deep-dive-into-machine-vision-in-safety-process/
[20] https://www.sciencedirect.com/science/article/pii/S2095809923002862