W11_Hilal_Estimating Cost At Completion (Part 4)
Problem Recognition
Estimate at completion is really important to management in order to forecast the time and cost at project completion. These series of blogs were created to discuss each method individually and then in this blog will compare the results and select the best alternative.
Feasible alternative
The Guild of Project Controls Compendium and Reference suggest using three main forecasting/ estimation techniques:
- Independent estimate at completion (IEACs)
- Monte Carlo Simulation
- Best Fit
Analysis of Alternatives
a. IEACs: BAC, BCWP, and ACWP are used along with CPI and SPI values in order to calculate the EAC for different scenarios.
b. Monte Carlo Simulation: BAC is used as starting point and allow for random function with 0.5% difference from the original value. This is repeated 1000 times in order to generate an average, maximum and minimum values.
c. Best Fit: ACWP data points are used to plot three trendlines to provide three case scenarios.
The PRET formula is then used to find the mean, sigma, and variance which then is used to calculate the probabilities.
Acceptance Criteria
The acceptance will require the method to be easy to use, very close to ACWP line, and less difference to the baseline.
Acceptance Criteria vs Alternatives
The table below shows the P50 of each alternative against the original budget of the project which is $ 5,565.
Table 1: Three Methods vs Original BAC
Now we compare the three alternative to ACWP and see which method follows the ACWP line.
Figure 1: Three Methods vs ACWP Line.
From the table and figure above, it seems that Monte Carlo simulation gave the best estimate which closely follows the baseline and ACWP line.
Table 2: Ranking of Feasible Alternatives.
Selection of Preferred Alternative:
Table 2 suggests that the best alternative is Monte Carlo Simulation
Tracking/ reporting plan
Although we selected Monte Carlo simulation for this case, this analysis needs to be done again for different cases as the results may vary. The management, also, needs to choose a confidence level and base on that we should do this analysis. The next step now is to take one of the completed projects and run the analysis and compare the results from each method to what was actually achieved.
References
Stephen, J. (2017).W14_SJP_Forecasts Part 6. Retrieved from https://js-pag-cert-2017.com/w14_sjp_forecasts-part-6/
Guild of Project Controls. (2015). GUILD OF PROJECT CONTROLS COMPENDIUM and REFERENCE (CaR) | Project Controls – planning, scheduling, cost management and forensic analysis (Planning Planet). Retrieved from http://www.planningplanet.com/guild/gpccar/assess-prioritize-and-quantify-risks-opportunities
Zaiontz, C. (2013). Real Statistics Using Excel. Everything you need to do real statistical analysis using Excel. Retrieved from http://www.real-statistics.com/regression/regression-analysis/
National Defense Industrial Association. (2014). A Guide to managing programs using predictive measures. Retrieved from http://www.earnedschedule.com/Docs/NDIA%20Predictive%20Measures%20Guide.pdf
OK, seems you have corrected the error in this blog posting.
ReplyDeleteKeep on going......
BR,
Dr. PDG, Jakarta