Reference no: EM132438621

Topic: Adaptation of Bayesian Inference to forecast construction project duration

Overview

Forecasting time to completion (project duration) as accurateas possible at each time section, especially at the beginning phase of the project is important for construction project management.B. Kim proposed the Bayesian betaS-curve method (2009), which simulates project progress curve (so called S-curve) by beta function and it is to be updated based on the actual progress applying Bayesian Inference at each time section.In this task, aforementioned study by B.Kim (2009) is traced for further study in the future. This task requires knowledge of Bayesian Inference and Reliability theorem, and skill of MS Excel and MATLAB to perform curvature fitting (prior distributions of Beta-S curve Parameters) and Mote Carlo Simulation (posterior distributions of Beta-S curve Parameters).

Outline of the task

1. At first, please thoroughly read the reference 1002 and understand the study itself.

2. The main target of this task is to trace the numerical calculations (do the same things) included in the reference 1002.

3. Once you properly understand the study, please perform same calculations described in numerical example.

4. Curve fitting technique is required to obtain prior distributions for the parameters of beta-Scurve.

5. Monte Carlo integration technique is required for posterior distributions of the parameters.

6. Actual progress data is to be obtained by reading the value from Fig. 2 in the reference. The accuracy of the reading is not required. Demonstration of the calculation is most important.

7. Please do aforementioned calculations by MS excel first. Then please do the same thing by MATLAB.

8. Lastly please briefly summarize in report.

Article - Probabilistic Forecasting of Project Duration Using Bayesian Inference and the Beta Distribution by Byung-cheol Kim and Kenneth F. Reinschmidt