In the previous chapters we had covered a lot of tools and techniques that are used in Quantitative Risk Analysis. The last one in the list is “Modeling and Simulation” which is part of the Quantitative Analysis and modeling Techniques sub group.
Modeling and Simulation – Converts uncertainties into potential impacts on Project Objectives that are specified at a detailed level.
The definition may sound complicated. But, in simpler terms – We are going to try to understand the potential impact risks will have from the whole project’s perspective. Simulations provide a way to look at a risk at a specific point in time (In future, of course) in the project and help us to understand how that particular risk would affect our project objectives.
It allows a more detailed representation/understanding of the Risk because, it generates many possible scenarios so that we can monitor and examine the outcome. When we analyze the scenario and the outcome, we will get a good idea of what combination of values resulted in what outcome and this information can be extremely valuable while trying to come up with a response for the Risk under question.
Uses of Modeling & Simulation:
The following are some uses of Modeling and Simulation techniques:
1. Determine which activities are likely to be on the Critical Path. Remember the Critical Path that we try to find out once we create the Project’s Schedule using the Develop Project Schedule process?
2. Calculate the Probable Cost of the Project
3. Calculate the Probable Value of the Project, especially to the Customer
4. To perform Cost Benefit Analysis while deciding on a potential response. For ex: If you want to Crash your project schedule to shorten the overall schedule or to make up for lost time, you can use simulation to evaluate the probability of success
5. Calculate probable time to complete the Project or a Sequence of Activities.
There are many benefits of using this technique. But, the PMBOK goes a step further and explicitly states that this technique has much lesser chances of misuse than the EMV Technique.
Monte Carlo Analysis/Technique
The most commonly used technique under the Modeling & Simulation category is the “Monte Carlo Technique”. It is performed using Software to perform iterative Simulations. We use software to do this because of the complexities involved in the calculation and more importantly the iterative nature of the analysis.
The input values are chosen at random from a possible set of values (A Probability Distribution of values like Cost or Schedule parameters) with the idea of simulating possible/potential outcomes.
For ex: Let us say, we want to analyze a Project Network Diagram that we created during the Develop Project Schedule process. This analysis simulates many scenarios, determining all the existing possibilities. It can help us identify the Critical Path, the impact delays in one activity could have on the others or the overall project schedule, how much chances we have on completing the project work within the planned time etc. In short, it will help us identify the overall Risk we are carrying with us with respect to the Projects Schedule.
The output/result is typically displayed as a Probability Distribution Chart as well.
As suggested in the previous section, Monte Carlo Analysis is typically done to identify Cost or Schedule Risk. For cost risk analysis we will use Cost Estimates as input. Whereas, for schedule risk analysis, we will use network diagrams and duration estimates as inputs. Based on the outcomes we can determine the likelihood of reaching our target (Finishing our project within cost and on schedule)
Just before starting on Monte Carlo Technique, I referred to a point about PMBOK suggesting that Modeling and Simulation has lower chances of misuse when compared to EMV. Did you pause and think why? If you totally missed that statement in a hurry to finish this chapter, please go back and read it again and then think why they would possibly make such a statement…
Steps in Monte Carlo Analysis:
The following are the steps in a typical Monte Carlo Analysis:
Step 1: plug in the input variables like Schedule & cost variables for each work package (From the Work Breakdown Structure) or the specific Risk that you are evaluating
Step 2: Use the Software that has been specifically created/designed for this purpose and simulate the potential outcomes repeatedly
Step 3: The output and results tell us the impact each variable has on our project as a whole. Store the Results.
Step 4: The results are analyzed using EMV or Probability Distributions
Step 5: The results of the Analysis is displayed using a Cumulative Distribution. Remember Cumulative Distributions from a few chapters ago?
Did you figure out why PMBOK made such a bold statement? If you haven’t re-read either the section or thought through the reason, spend some time to figure it out. It is pretty straight forward.
In EMV Analysis, We (Humans) actually do the analysis and hence based on our individual selves, we could easily fudge with the analysis or the outcome of the analysis. Whereas, in Monte Carlo analysis, the computer is doing the simulation and hence chances of misuse are considerably reduced. You may argue that, you could fudge the inputs and have a negative effect on the analysis, but even if you do that, the effect you would have on the outcome of the analysis will not be as severe as it would be if you intentionally performed incorrect EMV analysis.
Prev: Decision Tree Analysis
Next: Section Summary