FW-HTF-P: Towards Preparing Future Machinists: Exploring Tacit Knowledge in Machining with Artificial Intelligence and Extended Reality

本课题研究如何获取和传递机械加工中的隐性知识。有经验的机械师拥有大量对他们的决策至关重要的默示知识，但这些知识是个人的，特定于背景的，因此很难获得，也很难形式化和与他人交流。这不仅使培训新的机械师变得复杂，而且随着二战后婴儿潮时期出生的经验丰富的机械师退休，可能会完全失去这种默契知识。为了应对这些挑战，项目组将调查在多大程度上以及如何在人工智能(AI)和虚拟/增强现实(XR)的帮助下识别、理解和转移机械加工中的隐性知识。该项目发展基金的工作将有助于更好地了解有经验的机械师的默契知识，以及如何将其应用于他们的工作。它还将产生识别默会知识和基于XR的培训系统和评估指标的人工智能技术的新想法；这些将形成未来研究的基础，支持美国围绕制造业准备劳动力的战略计划。该项目将扩大一支来自多学科背景的研究人员团队，包括工程学、学习科学、计算机科学、游戏设计/游戏化、心理学和劳动力发展。拟议的研究围绕三个基本推动力构建：(1)开发一条人-主体研究管道，并将其应用于关键利益攸关方，以了解隐性知识的类型以及隐性知识的获取和利用方式。(2)探索人工智能和XR技术来理解、识别和转移隐性知识。(3)建立认知模型，评估人类在XR训练中的表现以及训练的可信度和接受度。该项目团队将特别注意开发一些想法，使目前在该领域代表性不足的人群更容易获得机械加工职业，并与附近一所向聋哑人和重听学生教授机械加工的学校的合作伙伴合作。这项研究的总体目标是加深对机械加工隐性知识的理解，支持机械加工培训的新方向，同时开发可应用于机械加工以外领域培训的一般方法。该奖项反映了NSF的法定使命，并通过使用基金会的智力优势和更广泛的影响审查标准进行评估，被认为值得支持。

This project investigates how to capture and pass on tacit knowledge in machining. Experienced machinists have a great deal of tacit knowledge that is essential for their decision-making, but such knowledge is personal, context-specific, and consequently difficult to acquire, as well as to formalize and communicate to others. This both complicates training new machinists and risks losing this tacit knowledge entirely as experienced machinists born in the post World War II baby boom retire. To address these challenges, the project team will investigate to what extent, and how, tacit knowledge in machining can be identified, understood, and transferred with the help of Artificial Intelligence (AI) and Virtual/Augmented Reality (XR). The work in this project development grant will lead to a better understanding of experienced machinists’ tacit knowledge and how it is applied in their work. It will also generate new ideas for AI techniques that recognize tacit knowledge and XR-based training systems and evaluation metrics; these will form the basis of future research that supports U.S. strategic plans around preparing the workforce for manufacturing industries.The project will expand a convergent team of researchers from multiple disciplinary backgrounds, including engineering, learning sciences, computer science, game design/gamification, psychology, and workforce development. The proposed research is structured around three fundamental thrusts: (1) A human-subject research pipeline will be developed and applied to key stakeholders to understand the types of tacit knowledge and the ways in which tacit knowledge is acquired and utilized. (2) AI and XR techniques will be explored to understand, identify, and transfer tacit knowledge. (3) Cognitive models will be established to evaluate human performance in XR training and the credibility and acceptance of the training. The project team will pay special attention to developing ideas that can make the machining occupation more accessible to populations that are currently under-represented in this field, working with partners at a nearby school that teaches machining to deaf and hard of hearing students. The overarching goal of this research is to deepen the understanding of tacit knowledge in machining and support new directions for machining training, while developing the techniques in general ways that might be applied to training in domains beyond machining.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.