W25.0_NH_Schedule Risk Analysis Using Risk PERT Analysis and Monte Carlo Simulation

I. Problem Definition

The Business Development Division in an oil and gas company is assigned to develop a schedule risk analysis of a Tuban Project to figure out how much contingencies for project’s schedule. Schedule risk is usually ignored in cost risk assessments. More recently cost risk analysis has included attempts to represent uncertainty in time, but usually this analysis takes place outside the project schedule framework. Recently the tools have been available to include a full analysis of the impact of schedule uncertainty on cost uncertainty (integrated cost and schedule risk analysis). In this blog, Author tries to implement the integrated cost and schedule risk analysis on one of the company’s projects.

II. Feasible Alternatives

The steps involved in assessing the risk of a schedule are as follows:

1. Create a complete and quality-checked CPM network.
2. Develop three duration estimates (optimistic, pessimistic, and most likely) for each activity.
3. Identify a duration distribution method for all activities.
4. Compute the path distribution.
5. Evaluate the result (completion date certainty and high-risk paths)
6. Initiate status monitoring

In compute the path distribution, there are two tools will be used, which are PERT analysis and Monte Carlo simulation.

III. Development of the Outcomes for Alternative

Table 1 below contains the project base schedule, and Figure 1 is the CPM network.

Table 1. Project Base Schedule.

Figure 1. CPM Network Schedule.

From CPM network schedule, the project duration is 395 days, with the critical path is activities 1-2-3-4-5c-5d-5e. Beside critical path, there are 3 other paths which is 1-2-3-4-5b-5d-5e, 1-2-3-5a-5d-5e, and 1-2-3-6-5d-5e.

Table 2 contains three duration estimates, where a, b and m are optimistic, pessimistic, and most likely respectively. These data used in order to define expected value for the duration.

Table 2. “Three Duration Estimates” of the Project.

IV. Selection of a Criteria

The result of PERT Analysis and Monte Carlo simulation will be compared to each other. From that comparison we will select the worst case to accommodate the schedule contingency of the project.

V. Analysis and Comparison of the Alternatives

A. PERT Analysis

By using PERT analysis, the expected time, variance, and standard deviation value is obtained as shown in Table 3.

Table 3. Result of Expected Time and Variance by Risk PERT Analysis.

B. Monte Carlo Simulation

By using Monte Carlo Simulation, the expected time, variance, and standard deviation value is obtained as shown in Table 4 and 5.

Table 4. Result of Risk Duration by Monte Carlo Simulation.

Table 5. Result of Expected Time and Variance by Monte Carlo Simulation.

Figure 2. Result of Monte Carlo Simulation for Critical Path.

Table 6 below shows comparison of schedule risk analysis using PERT analysis and Monte Carlo Simulation.

Table 6. Comparison of Schedule Risk Analysis using PERT Analysis and Monte Carlo Simulation.

VI. Selection of the Preferred Alternative

By using management desired probability of P90, contingency for Schedule Risk using PERT analysis and Monte Carlo Simulation is obtained, as shown in Table 7.

Table 7. Comparison of Schedule Contingency using PERT Analysis and Monte Carlo Simulation.

Finally, we decided to use the worst case of 49 days (result from PERT Analysis) as schedule contingency of the Project.

VII. Performance Monitoring and Post-Evaluation of Results

It is necessary to conduct strict monitoring during implementation of the Project, to prevent the completion time exceed the schedule contingency.

References:

1. Sullivan, G. W., Wicks, M. E., & Koelling, C. P. (2012). Engineering economy 16th Edition. Chapter 12 Probabilistic Risk analysis, pp.546
2. Humphreys, G.C. (2018). Project Management Using Earned Value, Chapter 17, page 326 to 328, Fourth Edition, Humphreys & Associates, Management Consultants.
3. Guild Of Project Controls Compendium and Reference (CaR). (n.d.). Retrieved from http://www.planningplanet.com/guild/gpccar/conducting-schedule-risk-analysis
4. Guild Of Project Controls Compendium and Reference (CaR). (n.d.). Retrieved from http://www.planningplanet.com/guild/gpccar/create-logical-relationships-sequence-activities
5. Paterson, S. J. (2018, January 6). A comparison between 8 common cost forecasting methods. Retrieved from https://pmworldlibrary.net/wp-content/uploads/2018/01/pmwj66-Jan2018-Paterson-comparison-of-8-common-forecasting-methods-featured-paper.pdf
6. The University of Iowa. (n.d.). Normal distribution Applet/Calculator. Mathematical Sciencess–College of Liberal Arts & Sciences, The University of Iowa. Retrieved from https://homepage.divms.uiowa.edu/~mbognar/applets/normal.html
7. Wain, Y. A. (2014, August 1). W24_YAW_Schedule risk analysis. Kristal AACE 2014. Retrieved from https://kristalaace2014.wordpress.com/2014/08/01/w24_yaw_schedule-risk-analysis/
8. Wain, Y. A. (2014, August 4). W25_YAW_Schedule risk analysis (2). Kristal AACE 2014. Retrieved from https://kristalaace2014.wordpress.com/2014/08/04/w25_yaw_schedule-risk-analysis-2/
9. Wahyu, F. T. (2019, December 4). W8.0_FTN_Schedule risk analysis (1). OCTOPUS 2019. Retrieved from https://2019octopus.wordpress.com/2019/12/04/w8-0_ftn_schedule-risk-analysis-1/
10. Wahyu, F. T. (2019, December 11). W9.0_FTN_Schedule risk analysis (2). OCTOPUS 2019. Retrieved from https://2019octopus.wordpress.com/2019/12/11/w9-0_ftn_schedule-risk-analysis-2/