Home » Project Management, Project Scheduling, Risk Management

The Monte Carlo Simulation execution trial

12 July 2010 shim_marom

histogram1 The Monte Carlo Simulation execution trialOver a week ago I came up with an offer to provide readers with the unique opportunity to run their project plans through a Monte Carlo Simulation and provide them with a high level summary of the results.

The response exceeded my expectations and up until now I was approached by 30 readers to provide them with an assessment of their plan.

Having gone through this experience I thought it would be appropriate to summarize my findings from this exercise, without, obviously, revealing the individual nature of the plans that have come under my hands.

All plans provided were in a Microsoft Project format and none was larger than 500 lines. To my surprise, none of the plans provided had the basic scheduling distribution populated (Most Likely, Optimistic and Pessimistic estimates) for each of the tasks. In order to overcome this issue I have applied an across the board triangular distribution rule of 75%, 100%, 125% (i.e. assuming that the Optimistic Estimate will also be 75% of the Most Likely Estimate and that the Pessimistic Estimate will always be taken as 125% of the Most Likely Estimate).

Not surprisingly and as expected, in all cases, the 80% likelihood of finishing the project on or before a certain date was well after the plans’ deterministic date (i.e. the date predicted by the software as being the project’s completion date).

Not all project schedules provided had costing information provided but in those who had an expected project cost, this as well has shown a deterministic cost well below the 80% likelihood mark.

As I’m excited by what I’ve seen I intend to run with this activity for few more days so the opportunity to use this option is still open (but not for much longer).

VN:F [1.9.3_1094]
Rating: 0.0/10 (0 votes cast)
VN:F [1.9.3_1094]
Rating: 0 (from 0 votes)

Related posts:

  1. Project Risk Management and the application of Monte Carlo Simulation
  2. Monte Carlo Simulation for Dummies

3 Comments »

  • Shim Marom said:

    new quantmleap post: The Monte Carlo Simulation execution trial http://bit.ly/cusoxd #pmot

    VA:F [1.9.3_1094]
    Rating: 0.0/5 (0 votes cast)
    VA:F [1.9.3_1094]
    Rating: 0 (from 0 votes)
  • lmau lmaublog.blogspot.com said:

    Do you plan to check the accuracy of the monte-carlo simulation results compared to actual results of the different projects ?

    VA:F [1.9.3_1094]
    Rating: 0.0/5 (0 votes cast)
    VA:F [1.9.3_1094]
    Rating: 0 (from 0 votes)
  • Anonymous said:

    Interesting question. One of the issues with checking on the accuracy of a monte carlo simulation is that once the simulation is done it is expected that some level of corrective and mitigation actions will be taken. That in itself renders the analysis no longer valid. In general terms, performing this sort of analysis is constant Work In Progress. You analyze, take corrective actions, re-analyze, etc.

    Cheers, Shim.

    VA:F [1.9.3_1094]
    Rating: 0.0/5 (0 votes cast)
    VA:F [1.9.3_1094]
    Rating: 0 (from 0 votes)

Leave your response!

You can subscribe to these comments via RSS.