FW-HTF-P: Training an Agile, Adaptive Workforce for the Future of Manufacturing with Intelligent Augmented Reality

美国制造业劳动力的未来面临着一系列挑战：(1)由于婴儿潮一代的退休而导致工人短缺；(2)由于先进技术的引入而导致技能组合的转变；(3)年轻一代对制造业工作缺乏理解和吸引力。因此，到2030年，预计将有超过240万个美国制造业岗位空缺，预计美国制造业GDP将损失2.5万亿美元。增强现实（AR）最近被用于制造业工人的体验式培训和技能提升。事实证明，通过将指令与工人经验的时空对齐，增强现实可以将新员工的培训时间减少50%。然而，有证据表明，工人对AR支架的过度依赖会导致知识的脆弱性，并降低适应新情况的表现。该项目将调查增强现实是否以及如何帮助制造业工人在车间发展敏捷性和适应性，同时避免与依赖技术和扼杀创新相关的风险。一种新的智能增强现实系统将使增强现实指令能够根据工人的任务表现进行动态调整，并提高他们掌握组装和维护等复杂任务的能力。这项研究将服务于制造业劳动力快速和终身技能提升的国家优先事项，特别是代表性不足和服务不足的少数群体。一个由学习科学家、劳动经济学家、认知心理学家、计算机科学家和制造工程师组成的融合团队将调查三个基本研究重点：(1)未来工作：将对雇主技能要求的变化进行劳动力市场分析，以了解AR技术在美国的引入程度以及未来工厂工人所需的技能组合。(2)未来技术：智能AR系统将被设计出来，通过自适应的指令脚手架来理解、预测和指导AR支持工人的行为，以了解他们的表现和专业水平。(3)未来工人：将进行假设驱动的人类受试者研究，以了解适应性AR支架对工人绩效、认知负荷和学习的影响。本研究的首要目标是通过提高未来制造业工人将所获得的知识和技能转移到车间新情况的能力，来平衡他们的效率和创新。来自工业界、政府和学术界的专家将在一个多学科研讨会上齐聚一堂，阐明增强现实技术在培训未来劳动力和弥合制造业技能差距方面的潜力和风险。该奖项反映了美国国家科学基金会的法定使命，并通过使用基金会的知识价值和更广泛的影响审查标准进行评估，被认为值得支持。

The future of the American manufacturing workforce faces a perfect storm of challenges: (1) a shortage of workers due to the retirement of the Baby Boom generation, (2) a shifting skillset due to the introduction of advanced technologies, and (3) a lack of understanding and appeal of manufacturing jobs among younger cohorts. Consequently, over 2.4 million U.S. manufacturing jobs are anticipated to be left unfilled by 2030 with a projected cost of $2.5 trillion on the U.S. manufacturing GDP. Augmented reality (AR) has been recently adopted for experiential training and upskilling of manufacturing workers. AR is proven to reduce new-hire training time by 50% through spatiotemporal alignment of instructions with worker experience. However, evidence suggests that overreliance of workers on AR scaffolds can cause brittleness of knowledge and deteriorate performance in adapting to novel situations. This project will investigate if and how AR can help manufacturing workers develop agility and adaptability on the shop floor while avoiding the risks associated with dependence on technology and stifled innovation. A new intelligent AR system will enable dynamic adjustment of AR instructions to worker task performance and enhance their ability to master complex tasks such as assembly and maintenance. This research will serve the national priority for rapid and lifelong upskilling of manufacturing workforce, especially underrepresented and under-served minority groups.A convergent team of learning scientists, labor economists, cognitive psychologists, computer scientists, and manufacturing engineers will investigate three fundamental research thrusts: (1) Future work: Labor market analyses of changes in employer skill requirements will be conducted to understand the degree to which AR technologies have been introduced in the U.S. and the skillsets workers will need in future factories. (2) Future technology: An intelligent AR system will be devised to understand, predict, and guide the behavior of AR-supported workers through adaptive scaffolding of instructions to their performance and level of expertise. (3) Future worker: Hypothesis-driven human-subjects research will be conducted to understand the impacts of adaptive AR scaffolds on worker performance, cognitive load, and learning. The overarching goal of this research is to balance the efficiency and innovation of future manufacturing workers by improving their ability to transfer the acquired knowledge and skills to new situations on the shop floor. Experts from industry, government, and academia will be convened in a multidisciplinary workshop to illuminate the potentials and risks of AR technology for training future workforce and bridging the skills gap in manufacturing.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.