The Quantitative Safety Assessment and Evaluation for Safety-Critical Computer Systems

Abstract

The rapid increase in computer-related technologies puts greater demands on the software-controlled systems. These demands place the software in total control over safety-critical systems. A fault in such a Safety-Critical Computer Systems (SCCS) can result in catastrophic concerns such as loss of life, harm, or equipment damage. In order to detect and prevent such faults, some safety standards, safety analysis and fault-tolerant techniques have been developed. Still, it requires adequate methodologies and frameworks to complete verification of these SCCS. In this work, the software safety is considered into 3 parts: Safety Modeling, Safety Measurement and Safety Management, with assurance lying down to develop high quality software for SCCS. The proposed methodology is SM-Cube (Safety Modeling, Safety Measurement and Safety Management) which is the Safety-Critical Computer Systems concerned. SM-Cube provides a clarified differentiation for assessing and evaluating the safety embedded into the software. SM-Cube consists of required processes and operations for developing SCCS free of faults. This proposed SM-cube has been applied to safety-critical software based Railroad Crossing Control System (RCCS) which is a laboratory prototype. The outcomes of the prototype confirmed that all critical operations have been safe and risk free. The development of the software based on the proposed SM-cube for RCCS have shown a simplified and improved safety-critical operations of the overall system performance