Week 3 Discussion - Quantitative Risk Analysis and Modeling Techniques

99 unread replies.99 replies.

This week we focused on risk. In the field of project management, there are a number of Quantitative Risk Analysis and Modeling Techniques used by project managers to determine the level of influence of the risks identified. We use these techniques because project managers are at the center of a long list of stakeholder questions, for which, PMs need to have answers. Some questions include:

• When do you reasonably expect to complete your project?
• Whats the probability of completing the project on time or on a given date?
• Which activities on the critical path should you focus your attention on to meet the schedule?
• Can you tell me with a 95 percent confidence level how much the project will cost?
• How much variance is associated with the total man-hours youve estimated for this project?

So, as project managers, we need to be prepared at all times. Lucky for us, there are a number of techniques to help us answer these and other questions that become part of our daily routine. We will focus on two for the purposes of this course:

The Three Points Estimation Technique

Two popular formulas:

1. Triangular distribution:

Triangular Distribution: E = (o + m + p ) / 3

where E is Estimate; o = optimistic estimate; p = pessimistic estimate; m = most likely estimate

2. Beta (or PERT):

Beta Distribution (PERT): E = (o + 4m + p ) / 6

The beta distribution is a weighted average in which more weight is given to the most likely estimate. This alteration to the formula and placing more weight on the most likely estimate is made to increase the accuracy of the estimate by making it follow the Normal Distribution shape. Hence, in most of the cases, the Beta (PERT) distribution has been proven to be more accurate than the 3-Point triangular estimation.

Sensitivity Analysis

Sensitivity analysis provides a quantitative measure of changes in project variables and their associated costs to the total project. Do you think a simple and minor change in a project has minor financial impact? In most cases, think again! Sensitivity analysis also provides the financial stakeholders (portfolio managers, CFOs, other decision-makers) insight into how sensitive the financial aspects of a project are when specific parameters are manipulated. One such financial indicator is the contribution margin, which--in the context below--refers to how a project contributes to the potential profits or cost savings of the organization.

For This Discussion

Answer both questions, using each technique described above.

Scenario 1

A widget manufacturer is considering two projects that impact several IT systems within its complex supply chain. Each project requires an initial $250,000 capital investment. Project A has an expected contribution margin of 40 percent and Project B has a contribution margin of 55 percent. The portfolio manager performs a sensitivity analysis on the two potential projects as illustrated here:

As this figure illustrates, Project A--with an estimated 40 percent contribution margin--has a healthy NPV. Recall from your accounting courses that NPV represents the value of future dollars at their present dollar value. Project B shows an even greater NPV value. A project candidate with a positive NPV is a likely candidate for selection. (source (Links to an external site.))

Question: If costs fluctuate (which they often do), which project is more responsive to risk? Why?

Scenario 2

You are the lead project manager for a new multi-million-dollar building project in downtown Chicago. The project is in the heart of the financial district and has disrupted the flow of commerce and traffic for a number of months now. Time is starting to take its toll. You're in a project update meeting with the CEO of the real estate developing company, and the Mayor of Chicago. Both are asking you how long the giant crane on LaSalle Street is going to be there because it is blocking 1.5 lanes of traffic.

Optimistic estimates from your team are 6 days, pessimistic estimates are 14 days, and the most likely estimate is 10 days.

To Do: Using a Triangular or Beta distibution, calculate the estimated days (E).