IEEE Electrification Magazine - September 2017 - 53

removing the potential misinterpretations among designers
they have a relevant effect on weights, volumes, and costs)
and analysts, making it possible to apply a unique reading
and are thus removed. In addition, it is possible to find comof the verbal requirements. Moreover, through its concepts
ponents that may be either improved, to enhance the whole
and techniques, functional safety at first enables the defiship's dependability (by substituting them with more
nition of an acceptable level of safety for the system in the
dependable but costly ones), or worsened, if their impact on
course of design, based on historical data and systems
global indexes is risible (by installing less dependable but
already approved, and then it allows evaluation of the
cheaper ones).
degree of closeness of all the possible design solutions to
Finally, it is remarkable that some marine applications
the defined target.
that use concepts similar to the ones already present in
Along with the significant improvements to safety that
SRtP exist. The ships endowed with DP systems must comcould be reached by using functional safety, the application
ply with several requirements concerning their response to
of dependability theory can provide several other advantagfaults and possible accidents. Such requirements are siges. Indeed, dependability techniques can be used to pernificantly stricter than the SRtP ones, especially for the
form optimization and refinement both in design and
higher DP classification levels. Indeed, the DP systems with
verification processes, along with common design drivers'
the highest performance (e.g., the American Bureau of
evaluations. In the following, some examples of such capaShipping DPS-3 class) shall continue to operate after a
bilities are given.
given casualty (fire and flooding, equally to SRtP), assuring
Regarding the aforementioned space-by-space verificathe same performance level as before the event (delivery of
tion approach, its base concept is similar to the FMEA prothe correct service). In contrast, SRtP regulation allows a
cess. Indeed, both are aimed at assessing the effects on the
certain degradation of the service, given that it is
overall system of an event: a single
component fault in FMEA, a single
space, or an entire MVZ in a spaceDesigners'
by-space approach. Moreover, both
Expertise
IMO MSC
imply evaluating the inputs and
SRtP
Design
Concepts
outputs of the system in the course
Process
of study, trying to find possible
causes and effects of faults in
Verbal
Numerical
Design
CS
System
FMEA, or assessing functional relaDesign Redesign
Requirements Requirements
Drivers
Interpretation
tionships in SRtP verification. Due
to such a similarity, it may be possiVerification
ble/useful to introduce into the SRtP
Process
verification process tools originally
developed for the FMEA (e.g., softAnalysts'
Approved
ware and modeling) to take advanExpertise
Design
tage of what has been already
developed for such a technique.
Figure 13. The current state of the relationships among design/verification processes and SRtP
In addition, the input-output
requirements.
relationships assessed by the spaceby-space verification can be used to
execute an FTA or an RBD, with a
IMO MSC
reduced effort with respect to a
SRtP
standalone analysis. Conversely,
Concepts
Design
Functional
such techniques may also be
Process
Safety
employed during design, thus
allowing an easy final verification to
Design
Numerical
CS
System Redesign
Verbal
be performed. Both techniques can
Interpretation
Design
Drivers
Requirements
Requirements
conveniently identify the single
points of failure in the system,
Verification
which are solvable through the
Process
application of proper design criteria.
Moreover, the excessive application
Approved
of redundancy criterion can also be
Design
indicated (redundant components
whose effect on the overall system's
Figure 14. The state of relationships among design/verification processes and SRtP requirements after dependability theory application.
dependability level is risible, but
IEEE Elec trific ation Magazine / S EP T EM BE R 2 0 1 7

Table of Contents for the Digital Edition of IEEE Electrification Magazine - September 2017