New Zealand Engineering 1997 June

Construction

Quantitative Risk Analysis - A New Tool for Steep Site Decisions

What is QRA
Example
Benefits and applications

Before issuing building or subdivision consents for sloping land, local authorities need to know how stable that land is. New techniques and guidelines are now available, according to a new report recently published by BRANZ, to help geotechnical practitioners quantify the risks inherent in any particular site. These assessments can, in turn, be included in their reports supporting subdivision and building consent approvals.

Since 1992 BRANZ has been undertaking annual surveys, across all sectors, to identify key areas in which the industry requires more research to be carried out. In the area of geotechnical research, slope stability has continued to be the topic attracting the greatest response. The work commissioned by BRANZ and outlined in the new report, SR 83 Quantitative Risk Assessment Methods for Determining Slope Stability Risk in the Building Industry, provides new tools to help meet this need.

Geotechnical engineers have, traditionally, relied on experience-based judgement to assess whether a slope is stable enough to allow development. This judgement was complemented by numerical techniques.

Now, however, new techniques are available. Both numerical methods and subjective judgement are used to quantify the risks inherent in any system, such as a slope, using Quantitative Risk Assessment (QRA).

What is QRA
QRA consists of assessing the probability of slope failure and identifying the consequences of failure. QRA is used to assess the risk of failure of engineered slopes - for example, cuts, fills and retaining walls, and of land that has already been affected by slope instability. It is more difficult to assess the risk of first-time sliding of natural slopes.

The actual process of QRA involves three steps:

1. Risk analysis involves assessing the types, characteristics and frequency of landslides in a given area, as well as considering the consequences of landsliding if people and/or property are impacted by failure. The output of this analysis is a mathematical expression of risk (a measure of the level of risk being analysed).

2. Risk assessment is designed to help define the limits of acceptable risk that are placed on members of the public, either as individuals or collectively, through exposure to landslides. The acceptable risks in different areas are considered, including loss of life and the economic, social and environmental consequences if failure occurs.

3. Risk management involves considering the analysis and assessment that have been made of the potential landslide and deciding how to control the resulting risk. Controlling the risk may involve mitigation, acceptance or avoidance.

Example

Figure 1 is a cross section illustrating a hypothetical example of QRA. In this case, an assessment is made of the specific risk of death to the occupant of a house that is subject to possible failure of a cut slope and foundation fill. Potential influences on the cut slope and foundation fill are shown. Actually calculating the specific risk is essentially a mathematical manipulation of the probability of failure, the elements at risk and the consequences of failure. These mathematical manipulations are explained fully in the study report.

Stability analyses indicate that, given a 50-year event occurrence, there is a 0.02 probability that the cut slope will fail. When this figure is combined with the probability of the element at risk (in this case a person) actually being impacted by the failure, the probability of that person being present at the time of impact and the probability of loss of life resulting from the impact of the failure, then a probability of 2 in 1000 is obtained.

In other words, the specific risk to life of the most exposed person in this house if the cut slope behind the house fails is 2 in 1000. This is the measure of the level of risk being analysed. By comparison, the specific risk to life if the fill in front of the house fails is only 3 in 10 000.

Once this analysis is made then a risk assessment would be carried out, for example by the consent authorities or the landowners, to assess whether this risk was acceptable to them. The final step is to consider how the risk might be managed or controlled.

Benefits and applications

The main benefit of QRA is that all cause and effect relationships associated with slope instability can be taken into account. This is not the case with conventional analyses, which emphasise the probability of failure rather than the consequences.

The application of QRA is limited by the difficulty of accurately determining the input parameters. However, the numerical expression of risk allows the best estimate of the stability of a slope to be communicated in terms easily understood by lay personnel. This means engineers, landowners, developers and regulators are able to understand the risks on which to base their development and planning decisions.