Civil Engineering Reference
In-Depth Information
For engineers working under other code regimes, the same
logic applies but will necessarily depend on the extent to which
risk concepts have been worked into the code requirements.
and explicit risk management methods, such as those described
in the following sections.
3.10.1 Methodology - hazard identification and logging
For any engineer carrying out explicit risk management,
there is no substitute for maintaining a Hazard Log (or Risk
Register, the terms are effectively interchangeable). A Hazard
Log summarises:
Box 3.3
Hidden risk factors: tolerances, workmanship
In general, deviations in dimension and quality - within the bounds al-
lowed by normal specifications - are taken as covered by load factors.
However, the adverse effect of tolerance on structural performance is
rarely if ever covered explicitly.
For example (from a recent investigation) the quoted tolerances on
levelling the subgrade below a ground-bearing slab and on levelling
the top of the slab resulted in the slab itself not having the intended
thickness and being inadequate for the intended loading. A deliber-
ate policy by the contractor of reducing the slab concrete in favour of
cheaper fill was suspected as part of the problem, but the combina-
tion of tolerances ('permitted deviations') had ultimately produced an
inadequate structure.
It is valuable to check, during the design, that the structural form,
details and performance are not critically dependent on close con-
trol of dimension or quality. When such control is needed, the design
information and specification are the means to manage the risk, al-
though it may also be appropriate to make the design more robust to
reduce the sensitivity to error.
The causal factor generating risk
The scenario in which failure and/or loss occurs
The measures in place which restrict causation or loss
The type and size of loss
The measures of likelihood, loss and - in combination - risk
Intended action if the risk is to be reduced
Hazard identification (Hazid) is the process by which the
Hazard Log is first populated. It may be a simple process of
running through a check list based on experience, which can
be done by one or two people. In more complex or sensitive
structures, it is sensible to mobilise the experience and imagi-
nation of several people in a workshop; the check list becomes
an agenda for the workshop.
Either way, the Hazard Log is less important than the process
of questioning the normal assumptions to determine whether
there are unusual risk issues or to confirm that conventional
design and analysis will be enough. (This is written in terms of
structural design but the process holds good for the CDM risk
register or a project time and cost risk exercise.)
A graphic matrix of likelihood and consequences is com-
monly used to put a coded risk marker in the hazard log. In
Figure 3.1
, 5 is the high end of the likelihood or consequence
scale, 1 is the low end. So 5 in likelihood might represent
'several occasions in the life of the structure' and 5 in con-
sequence might represent 'collapse or 100% loss of use for
several years'.
3.10 Innovation
Innovation can take the engineer into the realm of 'unknown
unknowns'. History shows that in some cases, the adoption of
logically justified forms or details gave rise to issues of under-
standing which had not previously been significant - shear
buckling in box girders, for example. On the other hand, the
widespread workmanship problems in precast concrete panel
systems around the same time might have been envisaged with
imagination at design stage.
By their nature, Codes of Practice cannot be expected com-
pletely to cover innovative structural forms. For that reason,
the engineers need to deploy both understanding of principles
Red -
unacceptable
risk
5
Green -
acceptable
risk
4
3
Blue -
acceptable
risk
2
1
1
2
3
4
5
Consequences
Figure 3.1
Example Risk Classification Matrix
