Quantitative Risk Analysis

Quick reference

• 00:04 Hi, I'm Ray Sheen.
• 00:05 Let's talk about doing quantitative risk analysis on project risks.
• 00:10 Quantitative techniques are often used with the high risks that you're
• 00:15 escalating, these risks often require major actions or
• 00:18 boundary changes to address them.
• 00:20 And the quantitative analysis provides an in-depth understanding of the impact that
• 00:25 risk will have on the project.
• 00:28 A word of caution,
• 00:29 I normally only do quantitative risk analysis on the risks I am escalating.
• 00:33 The analysis can take hours, sometimes even days to understand just one risk.
• 00:38 Therefore I don't bother with this analysis on small risks that the team is
• 00:42 handling, but only on the big ones that I need to talk with stakeholders about, and
• 00:46 that I want to have more analysis to support the discussion.
• 00:49 There are a number of quantitative techniques that you can use.
• 00:52 In my opinion, the best ones for
• 00:54 project risks is the sensitivity analysis, the what if analysis.
• 00:59 And that’s why I dedicated an entire lesson to that technique.
• 01:01 There are other quantitative techniques that I'll briefly discuss here,
• 01:05 including the failure mode effects analysis, PERT analysis, a decision tree
• 01:09 with expected monetary value analysis, and the use of simulations.
• 01:13 Let's talk about each of these in just a little more detail.
• 01:16 First, the failure mode effects analysis or FMEA.
• 01:20 The FMEA focuses on product or process risk, not project risk,
• 01:24 it analyses whether the results of what the project creates,
• 01:28 will work as expected by the customer.
• 01:31 The product or process risk is identified, and the severity or the impact
• 01:35 of that risk on the customer and the probability of it occurring are estimated.
• 01:39 Well so far, it's just like quantifying a risk matrix.
• 01:43 But now the special part of FMEA is that the ability to control or
• 01:47 detect the risk is also rated and quantified.
• 01:51 And then these three ratings are multiplied together to calculate a risk
• 01:54 priority number.
• 01:55 If that value is too high, a risk mitigation action must be taken.
• 01:59 While FMEA is great for product or scope risk, PERT is used for schedule or
• 02:03 budget risk.
• 02:05 The program evaluation and review technique, or PERT,
• 02:08 combines a weighted average of optimistic, pessimistic, and most likely estimates
• 02:12 when there's uncertainty with the individual activity estimates.
• 02:16 You can use PERT with either budget or duration estimates.
• 02:20 To determine the PERT, start with the network diagram or flow chart of
• 02:24 the project activities, and determine the three estimates for each task.
• 02:28 I will then calculate three project schedules and budgets.
• 02:31 One using all the best case estimates, one with all the worst case estimates, and
• 02:36 the third with the PERT estimates.
• 02:38 The PERT estimate is a weighted average of the best case, worst case, and
• 02:43 four times the most likely case, all divided by 6.
• 02:47 PERT is a good tool to inform stakeholders of the magnitude of uncertainty
• 02:51 on a project.
• 02:53 If you can take a risk response action to reduce uncertainty,
• 02:56 you can bring the best case and worst case closer together, and
• 02:59 improve your ability to manage the project.
• 03:01 The next technique is a decision tree with expected monetary value.
• 03:06 The project management body of knowledge, the PMBOK Guide, defines a decision tree
• 03:10 analysis as a diagramming and calculation technique for evaluating
• 03:14 the implication of a chain of multiple events in the presence of uncertainty.
• 03:18 So let's take a look at how it works.
• 03:21 By the way, this is a favorite technique of the PMP exam.
• 03:24 So if you intend to become a project management professional,
• 03:26 make sure you know how to calculate this one.
• 03:29 The decision tree diagram is normally combined with the expected monetary value
• 03:33 analysis, or EMV, it is a product of the probability of the risks,
• 03:37 times the monetary impact of that risk.
• 03:39 Let's look at an example.
• 03:41 We have a new product that requires an investment in our factory.
• 03:43 We need to decide whether to build a new factory, or expand our existing factory.
• 03:48 If the market demand is strong, we'll need lots of capacity.
• 03:51 If the market demand is weak, we obviously won't need as much capacity.
• 03:55 Marketing has told us that there's a 60% probability of the demand being strong.
• 04:00 Well let's consider our options.
• 04:01 We'll first look at the new factory option.
• 04:04 The new factory costs $120 million to build.
• 04:07 If the demand is strong, we'll generate about $200 million of profit.
• 04:11 And if the demand is weak, only about $90 million.
• 04:14 Now to calculate EMV for the strong demand.
• 04:17 The $200 million of profit is offset by $120 million of costs, for
• 04:21 a net of $80 million.
• 04:22 We multiply that times the probability of a strong demand of 60% to get EMV of
• 04:27 $48 million.
• 04:29 For the weak demand, the net is a minus $30 million, and we multiple that
• 04:33 times the weak demand probability of 40%, for a minus $12 million.
• 04:38 Now if we add up those 2 MB's for the build new plant branch,
• 04:42 we get a total of $36 million.
• 04:44 Now we'll consider using the upgrade branch.
• 04:46 Upgrading our factory means we'll still have to live
• 04:49 with some of the current factory constraints.
• 04:52 The profit for the upgrade branch is lower because of the extra cost and
• 04:55 constraints in the old facility.
• 04:57 The net of the strong demand on the path is only $70 million, and
• 05:02 the EMV is then $42 million.
• 05:05 While the net for the weak demand is a positive $10 million,
• 05:09 and the EMV is then $4 million.
• 05:11 That means that the EMV for the upgrade path is $46 million.
• 05:15 When we compare the two paths, we see that the upgrade path has the higher total EMV.
• 05:21 Even though there's less profit with the upgrade path under both conditions of
• 05:26 a strong demand, and a weak demand, the EMV shows us that in terms of total cost
• 05:30 to the business, the upgrade path is still better.
• 05:34 The final quantitative risk analysis technique is a project simulation.
• 05:37 Simulations are computer model of the project, all activities, resources,
• 05:42 constraints, uncertainties, and risks are modeled.
• 05:45 This modeling can take a long time, and it's inherently risky,
• 05:48 since if the model is wrong, the results will be wrong,
• 05:51 and there's no way to test the model without actually doing the project.
• 05:55 The model is then run using random combinations of all the uncertainties and
• 05:59 risk factors.
• 06:00 Usually done on a Monte Carlo simulation that creates a probability distribution.
• 06:03 In our illustration, the project could take anywhere from one month
• 06:07 to six months, with a 70% probability that it will finish in three months or less.
• 06:13 The impact of specific risks can be calculated by considering how
• 06:16 the distribution function changes when different risk responses are selected.
• 06:20 Take actions on the ones that resolve your biggest problems.
• 06:24 Whichever quantitative technique you choose, you'll be able to
• 06:28 assign a numeric value to the risk, instead of just red, yellow, or green.

Lesson notes are only available for subscribers.