AI accelerates an engineer’s innovation. It doesn’t replace them.
In the heart of every groundbreaking product, there lies a meticulously crafted set of requirements. Traditionally, this was governed by PLM specialists, business analysts and requirements engineers. However, in the age of AI, the landscape of requirement generation is undergoing a transformative evolution.
This post delves into the fascinating world where PLM specialists embrace the brilliance of AI-infused requirement generation, exploring its ability to accelerate processes while upholding the integrity of human craftsmanship.
AI in ALM and PLM: Where are We Now?
AI-assisted requirement generation in PLM and application lifecycle management (ALM) is not a replacement for human expertise. Supporters describe it as a symphony where human ingenuity harmonizes with the precision and speed of AI algorithms.
Detractors advise employees against entering confidential product information into public AI tools. The primary concerns are data security, results accuracy, potential misuse, regulatory compliance and social engineering attacks. The lack of control over data handling, limited understanding of chatbot security and the necessity to preserve an organization’s reputation all contribute to this cautious approach.
Fernando Valera, CTO at Visure Solutions, sees AI in PLM and ALM tools in another light. In the webinar “Unveiling V8: Revolutionizing Your Collaboration & Requirements Management Process with a new AI-Powered Version,” he said his company’s software interface draws inspiration from Gaudi, the Catalonia architect from Spain. Just as Gaudi found harmony in organic forms, AI algorithms can weave similar elegance into product requirement generation—creating new, previously unseen content. This is a good metaphor as AI is primarily based around learning from a given set of input data, understanding its patterns and generating new content that closely resembles or outperforms the original data.
How Systems Engineers and AI Navigate Complexity
Visure Solutions describes “requirements engineering [as] the process of defining, documenting and maintaining requirements in the engineering design process … This is about building the right system … a system that fits the user’s problems.” Requirement traceability includes continuous stakeholder communication and alignment. In other words, verifying and validating that target features and business values are approved, changed, managed and delivered to expectations. Every complex product and portfolio require advanced systems engineering tools and applications to enable accuracy and traceability across mechanical, electrical, software and recipe-based components.
Imagine a world where the tedious hours spent gathering, analyzing and documenting requirements are significantly reduced. AI achieves this by swiftly drafting detailed functional and non-functional requirements while ensuring a pace that matches the speed of innovation. For PLM specialists, this acceleration means more time for strategic decision-making, fostering innovation and diving deeper into the nuances of product development.
Systems engineers are the architects who guide cross-functional teams, and therefore AI algorithms, to ensure that requirements align seamlessly with the broader product vision. In this symbiotic dance between humans and AI, one can imagine that the final output has depth that pure algorithmic systems cannot achieve.
The power of AI-assisted requirement generation lies in its ability to navigate the labyrinthine complexity of modern product development. It goes beyond drafting; it needs to identify carryover requirements from one product variant to another, create a seamless transition between iterations, reduce duplication and foster reuse. This depth of understanding, coupled with the intricate nuances of industry standards and risk assessments, propels product development into uncharted territories of efficiency.
Embracing AI-Powered Requirements Engineering
In a post titled “Requirements Engineering: Step by Step,” Visure Solutions elaborated on the guiding principles and values of requirements engineering. It focused on five activities:
- Requirements must be defined and integrated across numerous stakeholder groups by domain, system and sub-system, regulation standard and more.
- Requirements must be analyzed and negotiated to refine user needs and constraints by clarifying the decision-making behind each tradeoff.
- Requirements must be documented and precisely specified to be locked, changed and managed through the engineering and manufacturing phases (and beyond).
- Requirements must be validated to ensure they are complete, concise and clear.
- Requirements must be continuously managed as a way of collecting feedback while analyzing and prioritizing all the products or requirements in the development phase.
Visure Solutions claims to have incorporated AI technology to enhance the quality analysis of product requirements. “By leveraging AI algorithms, the software automatically identifies potential errors or inconsistencies in requirements. This feature saves time and ensures that the specifications are comprehensive, accurate and aligned with industry best practices.”
The use of AI certainly goes beyond drafting initial requirements; it covers several activities, including:
- Gathering applicable systems standards, such as DO-178B/C, IEC 61508, ISO 26262, IEC 62304, FMEA and GAMP5.
- Creating initial requirement specifications, such as ISO 29148.
- Creating stakeholder lists and glossary terms.
- Extracting requirements from unstructured text data.
- Leveraging natural language processing (NLP) to identify ambiguous, missing, redundant or inconsistent requirements.
- Enforcing operational alignment and process adherence to support different models such as Automotive SPICE, CMMI, V-model, Agile or more.
- Generating test cases from requirement use cases and other assets of the projects.
- Automatically generating related requirements and identifying missing traces.
- Prioritizing requirements based on FMEA impact, supplier input and wider risks assessments.
- Improving team communication by incorporating role-based, capability-driven collaboration.
- Driving better product quality and reducing software defects.
Overall, the incorporation of AI into requirements engineering does not merely enhance processes—it creates a ripple effect across the entire product innovation ecosystem. It has the potential to ensure a seamless flow of information across teams. This interconnectedness transcends traditional barriers. It ensures that each element of the project is intricately woven into a cohesive whole.
Extending AI-Assisted Requirements Engineering across PLM
Now, these principles and expectations are not unique to ALM and software development; they indeed reach across the entire product innovation lifecycle and PLM scope. As businesses strive for market dominance and innovation becomes the lifeline of success, PLM specialists cannot afford to ignore the potential of AI-powered requirements engineering.
This synergy promises faster project completion, exceptional quality and a strategic approach to software development. In the dynamic landscape of PLM, precision and speed are pivotal; it leads to more efficient, accurate and adaptive product development processes. As such, AI is a strategic imperative for businesses aiming for excellence in the digital age. Responsively embedding AI into PLM processes will ensure effective human-in-the-loop integration at each stage of the product lifecycle.
In the symphony of requirement generation, AI is not a replacement; it’s a virtuoso collaborator, enhancing the skills of PLM specialists and elevating their craft. Together, they produce results that are not just efficient but deeply resonant with the essence of human ingenuity. By embracing AI assistants, PLM specialists can step into a future where their expertise is guided by the precision of algorithms.
