Towards an Automated Requirements-driven Development of Smart Cyber-Physical Systems

TL;DR

This paper explores automating the translation of natural language requirements into a formal model for developing smart Cyber-Physical Systems, aiming to reduce errors and improve validation.

Contribution

It analyzes the translation process from natural language to IRM-SA models and proposes automation methods using NLP techniques.

Findings

Identified steps in the translation process suitable for automation.

Proposed NLP-based methods for validating natural language requirements.

Enhanced understanding of requirements-to-model translation challenges.

Abstract

The Invariant Refinement Method for Self Adaptation (IRM-SA) is a design method targeting development of smart Cyber-Physical Systems (sCPS). It allows for a systematic translation of the system requirements into the system architecture expressed as an ensemble-based component system (EBCS). However, since the requirements are captured using natural language, there exists the danger of their misinterpretation due to natural language requirements' ambiguity, which could eventually lead to design errors. Thus, automation and validation of the design process is desirable. In this paper, we (i) analyze the translation process of natural language requirements into the IRM-SA model, (ii) identify individual steps that can be automated and/or validated using natural language processing techniques, and (iii) propose suitable methods.