As artificial intelligence advances manufacturing corporations, this evolution redefines both industrial business model innovation and reforms the manufacturing sector by using big data to drive the manufacturing process and associated decisions. One of the most promising approaches, Model-Based Enterprise (MBE), has shown its potential to drive smart manufacturing (or Industry 4.0) by linking all sources of digital data through the product lifecycle1. The global net value of the MBE market has grown from $7.89 billion in 20172 to $9.94 billion in 20193, and the forecast for the future market performance is set at about $44 billion by 2027. Beyond upgrading manufacturing equipment, companies have sought to develop a digital model-based network for higher production efficiency and a profitable return on investment. Unlike traditional manufacturing, the next generation of manufacturing networks will provide seamless product record-tracking and tracing capabilities for all parties, from customers to government regulatory compliance agents using machine learning (ML) techniques4, 5. The advances and implementation of MBE in engineering enterprises critically influence the practice of design. As MBE presents a unique opportunity to link all sources of digital data throughout the product lifecycle, we explore how the requirement domain can be linked to the CAD domain. In addition to engineers interested in machine learning implementations in product design, this research can benefit educators in developing ML models for ME students. This would allow engineering changes to be tracked both upstream and downstream for requirements and CAD analysis. For instance, design changes originated from requirements can be implemented in CAD, and vice versa. Further, it is important to consider how requirements and CAD can be visualized and realized during the early stages of the design process to help engineers reduce the risk of project failure. This is particularly pertinent as requirements often serve as the contractual agreement between parties, and thus all changes and decisions must be aligned with the corresponding requirements. However, this is difficult to perform as relationships between requirements and CAD are not formalized nor fully realized. Often, correlations are manually determined by experts based on their heuristic knowledge. By automating this process, engineers and designers would make AI-assisted decisions to provide better designs6.

In this paper, the purpose is to develop a framework for performing a study to address the requirement analysis challenges associated with engineering education in building digital threads for Industry 4.0. Digital threads in manufacturing can be divided into four domains: design requirements, CAD, computer-aided manufacturing (CAM), and quality inspection7. Tracing digital information across domains presents unique challenges in complex systems, primarily due to the high volume, complexity of requirements management, and the difficulties in interpreting them resulting from change propagation. Current CAD education primarily focuses on teaching low-level skill sets, whereas Industry 4.0 engineering would require the ability to combine domains such as requirements-CAD and CAD-CAM. Engineering design changes are often derived from requirements documents and propagated to CAD and CAM systems. However, it is also important to emphasize the importance of back propagation of information in design education. As an example, engineers must assess the compatibility of new design parts with existing design requirements efficiently in order to streamline the future design process …