Safety Function Model for Requirement Specification in Critical Systems: A Case Study of Generic Patient Controlled Analgesia Pump Model (CGPA)

Abdullah, Azma
Abu Bakar, Rohani
Abdul Farid, Fairus
Abdulhak, Mansoor

International Journal of Engineering, Science and Information Technology
Vol 5, No 3 (2025)

Developing safety-critical systems (SCS) involves a systematic method for assuring and providing safety and dependability. Conventional approaches rely on expert intervention, which can introduce bias, cause delays, and promote inconsistency. This work proposes a model that enhances efficiency and accuracy by extracting safety functions from requirements specifications. The model is made up of three main steps: (1) preprocessing, which involves getting rid of stop words; (2) string selection and matching using a database of safety properties variables based on literature and expert knowledge; and (3) putting safety and non-safety functions into a structured safety function log. The model was trained and tested with the CGPA insulin pump and got a 94% F1 measure score, which means it was 91% accurate, 96% accurate, 92% precise, and 96% recall. This shows that it is good at making things clearer and less biased when finding functions for safety against failures, malfunctions, operational hazards, and inconsistencies in safety-critical specifications. All these enhancements contribute towards Sustainable Development Goal (SDG) 11: Sustainable Cities and Communities, aiming to develop safer, resilient, and sustainable infrastructure in safety-critical regions.