Evolution of geotechnical risk analysis in North American practice - Risk and Reliability in Geotechnical Engineering - page 471

Figure 12.6 An example of a risk matrix or “heat” diagram.
12.8.3 risk registers
One of the advances to be reinforced in New Orleans and California was the use of risk reg-
isters. Risk registers are not new: Anyone who has ever written down a checklist of things
that could go wrong in a project has created a risk register. But modern systems risk assess-
ments take the risk register to a new level of detail and comprehensiveness. Such registers
are now common on large geotechnical projects informed by risk management, particularly
tunnels, to aid in informed risk management.
The risk register typically structures things that might go wrong on a project as rows in
a spreadsheet. These are usually organized by the asset type, or by phases of the project, or
in some other orderly way. The columns in the spreadsheet provide a description of the risk
item to a level of detail that two engineers reading the same description take away the same
understanding of the event, an appraisal of the probability and consequence should the risk
occur, and usually an assignment of responsibility to someone within the organization for
monitoring the risk item. Sometimes, the risk register also lists response and remediation
action plans for each significant risk item.
It is increasingly common to color code the risk items in the register on a simple ordinal
scale—low-medium-high—as shown in Figure 12.6 . This has proved to be a useful way
for grouping risks and tracking those with the greatest expected consequence and for com-
municating risks to project stakeholders. However, this practice is not without its critics
and suffers limitations for which account should be taken (Cox 2008, 2009). The issue with
risk matrices is that the rankings are ordinal scales and thus do not admit the mathematical
operations of addition and multiplication; thus, weighted sums of these measures or even
comparisons along diagonals through the matrix are not meaningful, despite that many
people overlook this limitation.
12.9 eMergIng aPProaCheS: SYSteM SIMulatIon,
StreSS teStIng, anD SCenarIo aPPraISalS
While geotechnical risk analysis has matured over the four decades since 1970, it is still
evolving. An argument could be plausibly made that the rate of advance is greater now than
ever before. Certainly, its reception in mainstream geotechnical practice is now an accom-
plished fact.