PROFES 2025
Mon 1 - Wed 3 December 2025 Salerno , Italy
Mon 1 Dec 2025 15:05 - 15:30 at Sala degli Affreschi (Fresco Room) - QuEMaLeS Session 1

This paper has the goal of investigating the degree of alignment and complementarity between AI-FSM (an AI-oriented Functional Safety process model) and the MLE processes integrated in ASPICE PAM 4.0. The paper examines the degree of alignment of two standards/models as deployed in the demo of a safety-critical ML system in the railway domain. The mapping of their lifecycle shows a strong compatibility, with some warning and limitations. The high-level process correspondences are useful, but this is not sufficient to ensure comprehensive alignment, and for this reason, a deeper mapping effort is required, covering practices, sub-practices, and work products content. The SAFEXPLAIN railway demo provides a real case study of the application of ASPICE MLE in a safety-related context. The project outcomes validate the applicability of the AI-FSM in an ASPICE context.

Mon 1 Dec

Displayed time zone: Amsterdam, Berlin, Bern, Rome, Stockholm, Vienna change

14:00 - 15:30
14:00
10m
Day opening
Workshop Introduction
Quality Evaluation of ML-based Software Systems
Giuseppe Lami Institute of Information Science and Technologies "A. Faedo" - CNR, Francesco Merola ISTI-CNR, Luigi Buglione DXC Technology, Fabio Falcini
14:15
25m
Paper
A Survey of Existing Standards Addressing AI-based Technologies
Quality Evaluation of ML-based Software Systems
Giuseppe Lami Institute of Information Science and Technologies "A. Faedo" - CNR, Francesco Merola ISTI-CNR
14:40
25m
Paper
Critical Analysis of ASPICE® 4.0 Machine Learning Engineering Process Requirements
Quality Evaluation of ML-based Software Systems
Fabio Falcini , Giuseppe Lami Institute of Information Science and Technologies "A. Faedo" - CNR
15:05
25m
Paper
Alignment and complementarity between AI-FSM and ASPICE MLE: findings from the assessment of the SAFEXPLAIN Railway Demo
Quality Evaluation of ML-based Software Systems