China
Africa
Explore chapters and articles related to this topic
Improved subgrades
Published in Burt G. Look, Earthworks, 2023
The statistical considerations for this case study are shown on the PERT probability density function (PDF) in Figure 12.7-3. The PDF outputs were used as inputs in the Latin Hypercube simulation model. While one can select one of several distribution functions, the normal distribution is generally not appropriate when analysing natural soil and rock data (Look, 2015). The PERT distribution is widely used in risk analysis to represent the uncertainty of the value of some quantity where the reliance is on subjective estimates because the three parameters defining the distribution are intuitive to the estimator. The PERT PDF requires the minimum, most likely maximum, and the static value to be specified. A Latin Hypercube sampling is then used in the simulation as this stratifies the input probability distribution as compared to the Monte Carlo sampling. The procedure is described in the Palisade Corporation @Risk Manual.
Randomness
Published in Raymond J. Madachy, Daniel X. Houston, What Every Engineer Should Know About Modeling and Simulation, 2017
Raymond J. Madachy, Daniel X. Houston
The PERT distribution is simple to implement and can approximate a symmetrical normal distribution or a skewed distribution like the triangle or lognormal. It is available in many packages and often gives good enough results given the inherent uncertainties. But if more precise information is known about a parameter distribution, then one should consider an alternative if the PERT is not a good fit and modeling precision is important.
Estimating the critical chloride threshold of reinforcing steel in concrete using a hierarchical Bayesian model
Published in Sustainable and Resilient Infrastructure, 2019
Note that several of the selected studies from the literature only reported an interval of the measured CT values without providing any other information regarding the number of tested samples or measures of central tendency and variability of data. In such cases where an interval of possible values is known, but the measurements provide little additional information regarding the location or scale parameters, a continuous uniform distribution, or PERT distribution, can be used for simulating pseudo-random numbers (DeGroot & Schervish, 2012). The PERT distribution that is extensively used in modeling expert estimates can be used if information about the most likely estimate (i.e. mode) of data are available. One advantage of the PERT distribution is its smooth shape that fits well to the lognormal distribution, which is a proper distribution for nonnegative data. Furthermore, as opposed to the uniform distribution that treats all the data points within a range equally likely, the prior knowledge about data can be translated into the PERT distribution by specifying the most likely value of the distribution. In this paper, a hierarchical data simulation technique is applied to simulate the CT data from the PERT distribution. The PERT distribution is a special case of the beta distribution that takes three parameters given by PDF as follows:
An economic loss model for failure of sewer pipelines
Published in Structure and Infrastructure Engineering, 2018
Mohamed Elmasry, Alaa Hawari, Tarek Zayed
PERT distribution is considered as a special case of Beta distribution in which minimum, maximum and most likely values are assigned to the probability density function. Equation (21) (Vose, 2000) is used to calculate the mean value in PERT distribution: