This process is carried out as a supporting activity to Risk Analysis project management activity, to back your qualitative analysis of risks with some solid numbers. It may also be possible that certain risks could not be analyzed qualitatively and hence you use quantitative risk analysis to prioritize them.
In some cases this process is treated as optional, especially when there is lack of data, and after analyzing the risks qualitatively the team moves to the exercise of planning risk responses.
What is the difference between qualitative and quantitative risk analysis?
Qualitative risk analysis is more subjective in nature, based on facts and figures from previous experience. However, quantitative risk analysis produces statistical numbers for each of the risks, thus making it easier to prioritize them. This process is analyzes effect of risks on project objectives.
What do we need?
Risk management plan, cost management plan, schedule management plan – these are plans that help you assess risk impact on project objectives.
Risk register – contains all risks that need to be analyzed.
How do we do it?
- Interviewing is almost like a combination of expert judgment and three-point estimates we saw in Estimate Activity Durations or Estimate Costs processes. You talk to different people about a set of risks that they are knowledgeable about, and gather information about worst case, most likely and best case scenarios. Along with these record reasons for them. This information will help you define a budget range that helps dealing with the impact if the risk is materialized.
- Probability distributions are used to plot range of cost and schedule associated with a risk. This data can also be built from the three-point technique you use while interviewing people, and try to get a range of cost and schedule that is possible if a risk is materialized.
Once this data is collected you can draw one of the shape distribution graphs. Commonly used ones are beta distribution that uses two value parameters (alpha and beta), and triangular distribution which uses three parameters (most-likely, best-case, worst-case). Cost and time values are represented on x-axis and probability values on y-axis.
Exam pointer> You are not expected to know the formulae or plot the graphs on the exam. Exam expects you to know just the names of these tools.
Figure 1: Beta Distribution and Triangular Distribution
Risk analysis and risk modeling
These are used to analyze and model the risks based on the data gathered.
- Sensitivity analysis is very useful when you want to look at impact of the risk on just one of the project objectives, while assuming that there is no impact on the rest of them. This is a good way to see all risks with just one impact area and decide how risks need to be prioritized. For instance, just looking at cost impact of all risks will help you see how the budget is going to be distributed across categories of risks.
One such tool is a Tornado diagram, which is basically a type of bar chart, that gives a visual indication of risks.
- Expected monetary value (EMV) analysis is about coming up with possible scenarios to deal with a risk and assessing how much each of those paths will cost the project. Look at this post for a detailed look at Expected Monitory Value analysis.
- Modeling and simulation translate detailed uncertainties of the project into their potential impact on project objectives. Monte Carlo simulation is used to arrive at a likelihood of achieving specific cost or schedule targets. This technique iteratively computes the model several times from randomly selected input values.
As an example, for plotting simulation of coin toss –
Drawing a large number of pseudo-random uniform variables from the interval [0,1], and assigning values less than or equal to 0.50 as heads and greater than 0.50 as tails, is a Monte Carlo simulation of the behavior of repeatedly tossing a coin. (reference: Wikipedia)
What do we get?
As in qualitative risk analysis, the main output of this project management activity is updates to the risk register. Risks are easily prioritized using this numerical outcome. Any other supported points for reasoning the outcome are also recorded in the risk register.
Considering previous and this processes, we have seen how using abstract thinking and statistical tools the risks are assessed for probability of occurrence and impact on project objectives. The next step is to plan risk responses. Before that let us first look at detailed Expected Monitory Value analysis (EMV) – one of the tools from this process.