Polarion ALM & semantha: How AI unleashes the efficiency in requirements evaluation
Learn to drive a car again every day? Learn the same software again every day? Make the same mistake again every day? That doesn’t make sense, does it? Because once something has been learned, that knowledge can be accessed anytime it is needed. Simple as that! So why do users in Polarion manually evaluate stakeholder needs and customer requirements every day – even though the content is only slightly different from what they have already reviewed? The fact is: The knowledge is there, the insights are there – the treasure chest of existing requirements and all their valuable data is open. It’s just a matter of connecting it and grabbing what’s inside. The following sections show how users overcome the challenges of requirements evaluation with the help of AI, specifically through the semantha extension in Polarion ALM.
Facing the challenges
At first glance, there may be a logical explanation for processing new stakeholder needs and customer requirements in manual steps: They are usually worded differently. This makes automation difficult and results in having to read and review each one. However, there are a number of drawbacks to this approach, as well as additional challenges:
- Extensive effort: Manually reviewing and linking requirements is, above all, repetitive work that requires a lot of effort and is very time consuming. In addition, significant resources are needed to ensure traceability from start to finish.
- Time pressure: When suppliers receive a new product specification from their customers, the clock starts ticking. However, while turnaround times are getting shorter, requirements are becoming more complex. This makes it even more difficult to be the first to respond and stay ahead of the competition.
- Risk of human error: Humans are not machines. They make mistakes. Tasks such as requirements analysis are particularly prone to error. But when it comes to identifying high-risk issues, it’s important not to overlook critical details and to avoid costly mistakes.
- Lack of experienced professionals: The expertise needed for evaluation is often solely located in the heads of experts, who are becoming scarcer for demographic reasons. Nevertheless, it is important to ensure consistent high quality and minimize potential risks during the evaluation and bidding process.
Faced with these challenges, it’s obvious that handling requirements without automation is no longer an option. No worries: The semantha extension is here to help.
semantha – the basis for an AI-supported process
Whether it’s specifications, stakeholder needs, or customer requirements – wherever text-driven processes determine everyday work, semantha unfolds its full potential. The adaptive AI reads and understands large amounts of text – regardless of wording or language. It uses various AI technologies, such as Large Language Models, to create a so-called semantic fingerprint. This fingerprint captures the essential features or characteristics of the text and extracts its meaning. Without training or expensive fine-tuning, it is possible to quickly compare and analyze any kind of stakeholder needs or requirements on a semantic level. Requirements engineers thus save themselves tedious, manual tasks and can instead focus directly on the results.
Polarion and semantha – knowledge where it’s needed
While Polarion serves as a comprehensive solution for managing the entire requirements lifecycle and provides full traceability, semantha seamlessly integrates and offers the toolkit needed to turn existing requirements knowledge into valuable insights. Users can continue working in their familiar user interface and stick to common basic workflows, while semantha gives them an extra boost. The extension provides engineers with several key functions that not only increase their productivity, but also put them in full control of their analysis results.
Key features:
- Create a knowledge base: To leverage existing knowledge, users centralize data, such as stakeholder needs and customer requirements via the semantha extension in Polarion. This “library” forms the knowledge base and can be adjusted anytime as needed.
- Knowledge-driven analysis: Based on this predefined knowledge, new requirements can be compared automatically on a semantic level with previously evaluated needs and requirements. This can be different types of requirements-related content, be they product-, customer-, or version-specific. In Polarion, the semantha extension then presents matching previously evaluated requirements to the user – along with all related information and references.
- Evaluate and connect related objects: With these helpful match results, users have a shortcut when checking the new requirements. They can also easily update attributes, such as scope, priority, and comments based on the attributes of the matching requirement. This way, requirements can be enriched with all the necessary attributes and links with just a few clicks. In addition, the system displays the related work item in Polarion, providing clear reference points.
- Stay in control: Users are in full control and always have the autonomy to override semantha’s suggestions. They can review, approve, decline, and update attributes or perform linking manually. This ensures that corrections can be made by users whenever necessary, maintaining flexibility and accuracy.
- Get a report: To get an overview, users can generate comprehensive, detailed reports based on the data analyzed and the comparisons made. For example, after performing a version comparison, any new or differing stakeholder needs are highlighted.
Mit dem Laden des Videos akzeptieren Sie die Datenschutzerklärung von YouTube.
Mehr erfahren
When continually improving AI and a human in the loop work in perfect sync
Continuous improvement of results
semantha is designed to improve its knowledge through learning, rather than classic ML training mechanisms. By allowing users to make manual assignments or changes, and to accept or reject proposed requirements, the AI iteratively refines its understanding and updates its knowledge. In this way, the system stays in sync with users’ actions and decisions, assuring that new or updated information is taken into account in the future.
Data consistency
Users can also rely on a consistent and future-proof data integration as the semantha and Polarion data models are fully compatible. semantha achieves this by incorporating Polarion’s existing data structures. The extension can be viewed as an additional cog that operates within established workflows and standards and adheres to Polarion’s item approval process, promoting consistency and reliability.
Traceability
The semantha extension builds upon Polarion’s traceability features making sure that every update and change is properly attributed to semantha. Moreover, updates to the knowledge base are auditable and can be traced back to the original information in Polarion. Manual overrides of semantha’s results are tracked as well. This way, users not only seamlessly interact with the AI but also get the traceability they are used to from their current process.
Elevate efficiency and boost productivity
The combination of Polarion and the semantha AI extension represents a powerful synergy in the realm of requirements engineering. This integration not only enhances the efficiency of evaluating new requirements but also promotes a more informed and consistent approach to requirements management.
The extension helps to analyze, compare, and evaluate data efficiently based on knowledge from previous projects. Best of all, connecting attributes with just a few clicks increases productivity many times over. semantha helps reduce the risk of overlooking critical details and improves the overall quality of the requirements evaluation through consistency and reliability. Furthermore, by streamlining the evaluation process, requirements engineers not only save time that can be spent on more value-added tasks, they can also help to quickly create high-quality quotes for customers.
Good to know: Evaluating stakeholder needs with an AI boost is just the beginning. It is already possible to formulate questions or get a concise summary of the content, including references that allow the navigation to the origin of the corresponding information. With the help of LLMs, it will be possible to support many other process steps throughout the lifecycle in the near future, such as requirements or software code generation and quality checks. Close cooperation between customers and partners is needed in order to drive forward the development of modern AI-driven requirements management.
Getting started today
The semantha extension is designed to integrate seamlessly with Polarion’s infrastructure and ensures a smooth transition to AI without disrupting established workflows. Remember: The knowledge is there. The treasure chest is open. semantha helps you leverage that wealth of information.
Get in touch with us or visit the Polarion Extension Portal.
