# Project Risk Management

What is the likelihood that a given project will be completed on time, or on budget? If you can estimate probabilities for various individual tasks, it's possible to generate a model of a complex project and then use Stochastic analysis — Monte Carlo simulation — to understand risks in the project.

Here's how it works:

1. Construct a model of your project on the Project Outline tab by creating individual tasks and projects. A project is just a logical group for more than one task.

2. For each task, enter the expected task value, plus estimates for optimistic (lowest, or best-case) and pessimistic (highest, or worst-case) results. Add dependencies — you have to dig a hole before you can install the pool.

3. Then look at the Report tab to see statistics about your project and understand the risks involved.

Click one of the example projects below and then look at the various tabs to see how it fits together.

# Example Projects

The first example is a time-based project with complex dependecies. In this example we are interested in modeling the duration of a project and understanding how inidividual components contribute to the overall completion time.

The second example is a cost model with no dependencies. Here, we are interested in finding out the potential cost of a project, as well as the probability that the project will be completed on-budget.

Set the defaults for new tasks in your project. You can always change the setting for each individual task, but setting the defaults can save time when building the model.

 Distribution Sampling Estimate Type

# Number Format

Define how you would like to see numbers displayed (you can enter numbers in any format).

# Random Number Generator

Set a seed value for the random number generator and re-run the simulation model. This can be useful if you want to get reproducible results.

A seed value of zero (0) will use the current system time.

 Model type: Dependency-Based (Time)Cumulative (Cost)

 Top to BottomLeft to Right

 Report data:

# Project Statistics

Percentile ranges eliminate the tails (extreme values) to show likely outcomes. In the simulation, 80% of the results fall in the P80 range. 90% of results fall in the P90 range.

Please note: values in this section are truncated to two significant digits for readability.

# Contribution to Variance

Contribution to variance, or single-factor sensitivity, is the regression value of the individual task to the overall project result.

In other words, how much of the variance in the overall project can be explained by the variance in the individual task?

# Distribution of Outcomes

The distribution of outcomes is a histogram showing how often, during the simulation, a given outcome occurred.

The expected, optimistic and pessimistic values for the overall project are included if they are in range of the graph.

# Project Probabilities

The expected value for the project is calculated from the expected value for all tasks, taking into account dependencies.

The probability that the result is less than or equal to the expected value represents the likelihood that the project is completed on schedule (or before), or on budget (or for less).

 Data set: Project CalculationTask Sample Values