Practical crash duration estimation for project schedule activities

Abstract

The increasing demand in project acceleration in the construction industry has led to the development of a practical crash duration estimation model for activities. This study aims to develop a general model of crash duration (CD) estimation for project activities based on crash duration coefficient (CDC) and normal duration (ND). This study uses questionnaires to explore the judgment of 425 respondents regarding CD data. Three rules of thumb are employed: developing the 45 CDC’s distribution for each project activity; generalizing the most likely of the distributions; and developing a CD estimation model. The combination of the bootstrapping and Monte Carlo simulation techniques are used to generate population distributions of the sampling. The practical CD estimation model of (0.55 ± 0.1) ND is proposed to project an acceleration problem. The accuracy of the CD estimation is determined from the general CDC distribution value of 55% with a tolerance of approximately ±20% (0.1 ND/0.55 ND ≈ ±20%) is as a stochastic value. The CDC can be stated as a qualitative expression of a specific property of an activity in predicting the CD, and it plays an important role for CD accuracy and can identify risk-taking preferences. Using the risk-taking category in an interval scale is more logical for the qualitative assessment of the preferences of decision makers. CDC distribution models can provide additional information on certain conditions regarding the current CD practice, in which the contractors’ understanding of the distribution pattern could have distinct advantages over other contractors, who only depend on their experience and preferences.

Keywords

• Practical crash duration estimation
• Crash duration coefficient
• Normal duration
• Monte Carlo-bootstrap
• Project Schedule
• Decision maker preferences