More sophisticated vehicle features to improve user experience and safety in the Automotive Industry applications have significantly increased the number of Electronic Control Units (ECUs) over the years. These changes were followed by the increased complexity in technical solutions for Embedded Software and interfaces among the ECUs. In this context, Systems Engineering (SE) plays a key role to achieve robustness on definition and implementation of complex systems.The overall system is broken down into lower functions associated to a specific use case, considering user’s input and the expected outputs, in three different abstraction levels – Concept, Logic and Technology. Those functions are described with structure and behavior diagrams and requirements using SysML (Systems Modelling Language), which enables the entire system simulation and validation before being cascaded to ECUs Software implementation team during the Product Development process through Model-Based Systems Engineering (MBSE) tools.On this scope, the requirements have a vital role inside Systems Engineering methodology because it is the statement which defines the system required functionalities, performance, constraints, and other characteristics that satisfy stakeholders and costumers needs. The SysML diagrams are built based on the defined system requirements and many times this is a very iterative process, time-consuming and error-prone task until satisfying every required function logic and conditions. When the requirements are poorly defined, these can lead to project failure. Therefore, requirements must be unambiguous, verifiable, and traceable through Systems Engineering lifecycle. By using Easy Approach to Requirements Syntax (EARS) template, Systems Engineers have specific words to standardize the requirements, accomplishing with EARS requests such as “When,” “While,” “Where,” “If” to mention the stimulus or pre-conditions, “Then” to call the function/product and “shall” representing the action to be taken. In addition, by using the rules proposed by the International Council on Systems Engineering (INCOSE), Systems Engineers can improve the robustness of the requirement by avoiding, for example, words that may be misinterpreted during the system modeling and development process. In this context, Natural Language Processing (NLP) can be used to improve process flow by automating requirement verification. NLP techniques will be responsible for mapping and processing the text data in the system requirements based on actions, preconditions, functions, and triggers to improve the quality and robustness of the developed requirements.By Applying NLP algorithms, this paper aims to consolidate a more robust methodology to verify the system requirements, based on EARS template and INCOSE rules, prior to SysML diagrams development, which improves the final overall system implementation to satisfy the stakeholders and customer needs.


    Zugriff

    Zugriff prüfen

    Verfügbarkeit in meiner Bibliothek prüfen

    Bestellung bei Subito €


    Exportieren, teilen und zitieren



    Titel :

    Systems Engineering Process Enhancement: Requirements Verification Methodology using Natural Language Processing (NLP) for Automotive Industry


    Weitere Titelangaben:

    Sae Technical Papers



    Kongress:

    SAE Brasil 2023 Congress ; 2023



    Erscheinungsdatum :

    2024-01-08




    Medientyp :

    Aufsatz (Konferenz)


    Format :

    Print


    Sprache :

    Englisch




    Requirements Engineering for Automotive Embedded Systems

    Staron, Miroslaw | Springer Verlag | 2019



    From Natural Language to Semi-Formal Notation Requirements for Automotive Safety

    Krammer, Martin / Stirgwolt, Philip / Martin, Helmut | British Library Conference Proceedings | 2015


    From Natural Language to Semi-Formal Notation Requirements for Automotive Safety

    Martin, Helmut / Krammer, Martin / Stirgwolt, Philip | SAE Technical Papers | 2015


    From natural language to semi-formal notation requirements for automotive safety

    Krammer,M. / Stirgwolt,P. / Martin,H. et al. | Kraftfahrwesen | 2015