Paper Title
Modeling Method for Enhanced and Generalized Reliability Prediction in Safety-Critical Software with Simplified Model Selection

Abstract
Our proposed approach in this paper presents an integrated software reliability model (CSRM) that combines multiple software reliability evolution models (SRGMs) of the non-homogeneous Poisson process (NHPP) to achieve a balance between precision, adaptability, simplicity, and stability. By using machine learning, the CSRM efficiently combines the strengths of each SRGM while balancing their weaknesses through appropriate evaluation and calibration techniques. The developed CSRM has been successfully applied and validated to enable more efficient and accurate evaluation of software component reliability targets in autonomous robotic wheelchairs (ARWs). Based on the validation results, this novel approach has significantly improved the efficiency of the assessment process to achieve the defined reliability objectives. It also enables a more accurate assessment of the need for additional test executions and better planning of the required verification and validation sessions. This innovative approach provides valuable insight into the reliability of the developed software, especially for software developers who have limited experience in identifying and applying appropriate SRGMs. Keywords - Reliability Prediction, Safety Critical Software, Software Reliability Growth Model