Title: Software
1Software
Modelling Data Variability
2- 1GS314 RISK ANALYSIS ASSESSMENT
SCHOOL OF EARTH ENVIRONMENTAL
SCIENCES Practical 1 Bestfit (Granite Data)
3The Modelling Process
Data
FOS
Need to Describe VARIABILITY
gt
Model Equation
Monte Carlo Simulation
Statistics Distributions Fit a Variety
QRA
Test Goodness of Fit
Expert Judgement
4Software
5Granite Data
174 151 176 186 187 185 171 178 189 135 197 201 15
0 179 181 177 205 195 196 191 201
Geological Issues?
6Enter Data
7Distribution Specification
8Distribution Specification
9Run Fit Results Presented
10Run Fit Results Presented
11Run Fit Results Presented
12Comparison Graph
INPUT DATA
Comparison Graphs Input and result distributions
are superimposed on the same graph. Allows you to
determine whether the best fit distribution is
well matched to the input data in specific areas.
i.e. it may be important to have a good match
around the mean or in the tails.
FITTED MODEL
13Run Fit Results Presented
Statistical Distribution to Describe Variability i
n Data
14Difference Graph
Difference Graphs The difference graph displays
the absolute error between the input and results
distributions. This error is defined as the
difference between the input and results
probability for each input value of x. Comparing
the magnitude of the error to the magnitude of
the result gives an idea of the extent to which
the results deviate from the input.
15P P Graph
P-P Graphs Probability-Probability (P-P) graphs
plot the distribution of the input data versus
the distribution of the results. If the fit is
good the plot will be nearly linear.
16Q Q Graph
Q-Q Graphs Quantile-Quantile (Q-Q) graphs plot
the percentile values of the input distribution
versus percentile values of the results. If the
fit is good the plot will be nearly linear.
17Populations?
18135 MPa deleted
19151 MPa 150 MPa deleted
Uniform Distribution Bounded Between 169MPa
207Mpa
20Difference Graph
Is this better ?
21P P Graph
22Q Q Graph
23?
Original
Edited
Answer or Aid to Decision Making?
JUDGEMENT
24The Expert ?