[SystemSafety] New IET report

[Derek M Jones]

Peter,

> (4) develop a new standard for functional safety of programmable safety-related systems; and

Why is a new standard needed?  The conclusion states the reason
very clearly (i.e., the data is needed to apply the existing
standards is not available).

"The current approaches to describing safety in all four sectors are, to
some extent, empirical. These have been reasonable methods until now,
as advances in sectors have been somewhat incremental.

However, autonomy presents a complete paradigm
shift that is heavily dependent on recent advances
in data sciences (DS) and artificial intelligence (AI).
Many of the AI/DS technologies that contribute
to autonomy rely on scientific and theoretical
advances in these fields made in the last 10
years. Consequently, the field of verification and
validation providing system assurance has yet to
bridge the enormous semantic gap between these
new technologies and, to varying degrees, the
empirical models of safety employed in the four
transport sectors."

-- 
Derek M. Jones           Evidence-based software engineering
blog:shape-of-code.coding-guidelines.com