Using Metrics in the Industrial IoT Value Chain to Drive Trustworthiness
but are not enough to meet the needs of
managing trustworthiness and creating
business value. Business context and
information must also be considered.
Combining appropriate business metrics and
information
with
trustworthiness
understanding and metrics will allow an
organization to turn data into knowledge
that it can act upon to enable a dynamic and
successful enterprise.
measured by a metric involving the
percentage of overtime (OTpercent) for
processing of a production lot due to either
replacing, repairing or simply reusing a
defective machine:
ResACmetric = 100 - OTpercent
where a value of at least 80 is expected for
some types of failure.
Consider now a performance metric for this
assembly chain:
Organizations are generally concerned with
managing risks, both those associated with
trustworthiness aspects as well as others
(for example product delay or lack of market
adoption). A common approach to
measuring risks is to calculate the expected
value based on probabilities of events as well
as the anticipated impact of the event.
Leading the business while considering
business mission and goals, financial metrics,
risk
position
and
trustworthiness
considerations will require making complex
investment decisions involving many factors.
This
is
complicated
since
some
trustworthiness aspects will support a
business metrics and others will detract.
When all factors are considered together,
the tradeoffs will lead to an “acceptability
zone” where all objectives are reachable
(e.g., safety and performance). The details
depend on the precise definitions as
adopted by the business and industry in
question.
PerfACmetric = (expected processing
time of a production lot)/(actual processing
time)
where a value less than 0.9 in average is
considered unacceptable.
It is understandable that these two metrics
may depend on each other: the more
resilient the assembly chain, the greater the
certainty that its performance level will be
stable, according to these metrics. Figure 5
illustrates a strong dependency between
both, in case of hardship:
Trustworthin ess properties generally have
an impact on operations and business
outputs. Consider an IIoT system in a factory
that comprises an assembly chain. The
resilience of such a system includes the
resilience of its assembly chain. The
resilience of the assembly chain can be
- 94 -
IIC Journal of Innovation