Statistical-Based Metric Threshold Setting Method for Software Fault Prediction in Firmware Projects: An Industrial Experience

2026-02-01

Journal of Systems and Software

AI Summary

This study addresses the challenge of deploying black-box machine learning models for fault prediction in industrial embedded firmware, which often fail to meet functional safety standards due to their lack of interpretability. To overcome this limitation, the authors propose a cross-project method for establishing software metric thresholds based on statistical analysis. Specifically, C-language firmware metrics are extracted using Coverity and Understand tools, and hypothesis testing is employed to identify discriminative metrics and derive empirical thresholds. The resulting model requires no retraining or domain-specific fine-tuning and can be directly applied to independent projects. Experimental evaluation on three real-world industrial firmware projects demonstrates that the approach accurately identifies error-prone functions, enabling effective preventive quality interventions. Moreover, its interpretability and operational clarity align with stringent safety standards such as ISO 26262, offering a viable alternative for industrial software quality assurance.

Technology Category

• Machine Learning: Calibration & Uncertainty Quantification
• Natural Language Processing: Safety and Robustness
• Knowledge Representation and Reasoning: Diagnosis and Abductive Reasoning

Application Category

• Security and Privacy: Large-scale security measurements
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics

Abstract

Ensuring software quality in embedded firmware is critical, especially in safety-critical domains where compliance with functional safety standards (ISO 26262) requires strong guarantees of software reliability. While machine learning-based fault prediction models have demonstrated high accuracy, their lack of interpretability limits their adoption in industrial settings. Developers need actionable insights that can be directly employed in software quality assurance processes and guide defect mitigation strategies. In this paper, we present a structured process for defining context-specific software metric thresholds suitable for integration into fault detection workflows in industrial settings. Our approach supports cross-project fault prediction by deriving thresholds from one set of projects and applying them to independently developed firmware, thereby enabling reuse across similar software systems without retraining or domain-specific tuning. We analyze three real-world C-embedded firmware projects provided by an industrial partner, using Coverity and Understand static analysis tools to extract software metrics. Through statistical analysis and hypothesis testing, we identify discriminative metrics and derived empirical threshold values capable of distinguishing faulty from non-faulty functions. The derived thresholds are validated through an experimental evaluation, demonstrating their effectiveness in identifying fault-prone functions with high precision. The results confirm that the derived thresholds can serve as an interpretable solution for fault prediction, aligning with industry standards and SQA practices. This approach provides a practical alternative to black-box AI models, allowing developers to systematically assess software quality, take preventive actions, and integrate metric-based fault prediction into industrial development workflows to mitigate software faults.

Problem

• software fault prediction
• metric thresholds
• firmware projects
• interpretability
• industrial application

Innovation

• statistical threshold setting
• cross-project fault prediction
• interpretable software metrics
• firmware quality assurance
• static analysis