Oracle | Primavera Risk: Pertmaster Simulation Options

The sixth step of a typical Oracle | Primavera Risk (Pertmaster) Monte Carlo analysis is running the analysis.  This step is tougher on the computer than it is on a project manager.  This section will outline the risk analysis options that will be set prior to running the Monte Carlo simulations.

Video Walkthrough of Risk Analysis Options in Oracle | Primavera Risk (Pertmaster) (Duration: 13 min: 23 sec)

Risk Analysis Options Overview

The project manager will follow the menu path Risk | Risk Analysis Options.  The risk analysis options allow a risk analyst to control many of statistical options that will impact how the Monte Carlo analysis is run.

Risk Data Tab

  • Calculate Duration Sensitivity – the feature allows users to let Primavera risk automatically calculate the duration sensitivity upon completion of the simulations.  This is not a necessary option.
  • Calculate Cost Sensitivity – this feature is the same as the duration sensitivity option listed above.
  • Calculate Risk Percentiles – Primavera Risk can populate user-defined fields based on a desired confidence level.  For example, a field containing 80% confidence level dates or durations could be populated for every activity.  This will allow a scheduler to create probabilistic schedules if desired.
  • Save Resource Data – Pertmaster will record the resource usage for every iteration if this option is selected.
  • Save Probabilistic Cash Flow Data – this option will need to be selected in order to view pay period cash flow data.  Clicking the “Cash Flow” button will allow users to select which resources to accumulate data for.
  • Save NPV and IRR values – net present value and internal rate of return data can be viewed in the histogram based on the resource assignments.

Analysis Tab

  • Analyze for <x> iterations – the risk analyst can select the number of simulations to be run.
  • Use Latin Hypercube Sampling – Latin hypercube is a sampling method that converges in fewer iterations than a true Monte Carlo sampling.  The results should be basically equal.  Latin hypercube may be a good option for running quick scenarios in an interactive group.
  • Analyze until converged or maximum iterations are run – the simulation has reached convergence when the histogram curve changes plus or minus a set percentage over a pre-determined number of iterations.  For example, if a curve has changed less than plus or minus one percent over 200 iterations, then it might be considered converged.
  • Show step through analysis options – step through analysis options will allow users to view or pause the simulations to see the cost or schedule risk analysis in motion.
  • Set own seed for random number generator – setting a seed number will ensure that the random number generator will start at the same point each time allowing users to recreate a data set exactly.  If this setting is not selected, then Pertmaster will use the clock time to set a seed number.
  • Restore all ignored tasks for analysis – if any tasks have been ignored using the ignore task column, then this option will override the setting for the risk analysis.
  • Ignore correlation between risks – Pertmaster can turn off all correlation values for risk modeling or scenario testing purposes.
  • Resource level after each iteration – Primavera Risk has the ability to resource level during the iterations.  A schedule risk analysis with resource leveling may be incredibly time consuming.  The durations will be applied for each iteration and then activities will be leveled based on the resource leveling settings and priorities.
  • Show <chart name> after analysis – a chart of the user’s choosing will be displayed immediately after running the risk analysis.  The default option is the histogram chart.

Warning Tab

The warning listed below do not mean that the inputs are flawed.  They can be used as indicators that a mistake has been made.

  • Most likely risk distribution value is different from the deterministic value – this warning is an indicator that the remaining durations in the schedule are not considered to be most likely.  This may indicate that the values were entered incorrectly or some activities are not actually set to a “most likely” value.
  • Deterministic value is outside the risk distribution range – the warning is similar to the item above except it is more extreme.  The scheduled value is actually outside of the risk range.  This may indicate that the risk ranges were input incorrectly or that the project team has viewed the activity to be outside of the scheduled duration.
  • Risk is assigned to a task with actual finish – risk will not affect completed activities.  It is best to remove risk on finished activities for audit purposes, however it will not affect the risk analysis results.
  • A Beta Pert distribution is symmetrical AND a shape value has been entered – Pertmaster will ignore any shape input for a modified Beta Pert distribution other than the value of 4.