W13_Anton_Comparison PERT and Monte Carlo


W13_Anton_Comparison PERT and Monte Carlo

1.      Problem Recognition, Definition, and Evaluation

On blog W12.3, PERT method to define schedule contingency refer to risk identified, percentile desired is 90%.

This week, using the same data, I would like to apply Monte Carlo simulation as comparison to PERT.

2.      Development of the Feasible Alternative

Fajar (2008) stated: “ Monte Carlo simulation is a method for modeling and analyzing system which involving risk and uncertainty”. (p. 227).

And Sullivan (2012) said:  “ At the heart of Monte Carlo simulation is the generation of random numbers” (p. 503).

Using Ms. Excel to generate random number between zero to one by use function NORMSINV(RAND() to performing 1.000 simulation.

3.      Development of the Outcomes and Cash flows for Each Alternative

1

 2

3

5

4

7

6

8

Schedule P90 from Monte Carlo with total duration = 260.51 day

9

 4.      Selection of Criterion (or Criteria)

From W12.3 blog posting Plot duration P 90 to schedule from PERT result of duration are 285.64 day. And from Monte Carlo are 260.51 day.

5.      Analysis and Comparison of the Alternative

Both PERT and Monte Carlo method has different 25.13 days in result.

6.      Selection of the Preferred Alternative

Follow Humphreys (2012): “PERT technique is largely historical. Monte Carlo is understood best and is most frequently used” (p. 346).

7.      Performance Monitoring and Post Evaluation of Result

Also refer to Humphreys (2012): “ The biggest challenge remains, which is collecting the best data to use for any of these techniques” (p.346)

8.      References :

  1. Fajar, Adnan (2008), Aplikasi Simulasi Monte Carlo dalam Estimasi Biaya Proyek, Jurnal SMARTek vol.6, No. 4.
  2. Sullivan, William G., Wick, Elin M., Koelling, C. Patric. (2012), Engineering Economy. Chapter 6 Pp. 211 -228, Fifteenth edition, Prentice hall.
  3. Humphrey and Associates (2012), Project management using earned value , Chapter 27. Earned Value, Second Editions, Humpreys & Associates, Inc.
This entry was posted in Anton. Bookmark the permalink.

1 Response to W13_Anton_Comparison PERT and Monte Carlo

  1. drpdg says:

    WOW!!! Awesome case study and you did an OUTSTANDING job on your calculations and analysis, Pak Anton!!! Nice work!!!!

    About the only suggestion I can offer is to download the latest GAO “Scheduling Best Practices” http://www.gao.gov/products/GAO-12-120G and you can see how they recommend using simulation. What you might want to consider would be to take your schedule and benchmark it against the GAO “best practices”. Use that comparison for another blog posting?

    Another suggestion you may want to consider as a topic for future blog postings using this same case study would be to apply “Line of Balance” scheduling….. http://guidebook.dcma.mil/79/evhelp/lob.htm or http://www.cpmtutor.com/c02/lineofbalance.html Using the LoB method is PERFECT for your case study and I think it will prove very useful to you if you use your blog posting to explore whether LoB method will work better than Monte Carlo Simulation.

    I am also keen to see you follow up and let us know whether PERT or Monte Carlo proved to be the more accurate method…… Very interesting opportunity….

    BR,
    Dr. PDG, Jakarta

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s