Examples of the Implementation of Fuzzy Models in Tourism in the South Moravian Region - Geoinformatics for Intelligent Transportation - page 175

7.4 Comparison of the Methods and Defuzzi cation

The data of well-known parts of the bike trails which were possible to classify were
selected to compare the best methods. We can use the comparison of the maximum,
minimum, arithmetic mean and standard deviation according to the dif

culty of the

bike trails, then we monitored frequency histograms.
Mamdani cation CoS but also with

defuzzification CoM, where there is the bigger value range and the higher fre-
quency on the intervals of the maximum occurrence.
Larsen

' is method is well representative with defuzzi

is method has similar characteristics. However, it is not suitable for the
bike trails with the intermediate dif

'

culty and highlights the bike trails with the

small and hard dif culty.

The table compares the percentage of the bike trails suitable for the membership
in interval 0

h 25 ; 1 i according to small, intermediate and hard dif culty having

;

regard to their whole choice for the individual dif culties and the choosing meth-

ods. (Table 1 )
The sum value of the percentages expresses the precision of the individual
method. We can see that Mamdani

' s method bluntly dominates, especially with

defuzzi cation CoS.

The results of Larsen is method are quite good. This method is not much reliable

in the evaluation of the intermediate dif

'

cult bike trails. It signi cantly competes

with Mamdani cult bike trails. These

methods are compared at the selected region (Fig. 16 ), where Mamdani

' s method within the small and hard dif

' s method

with defuzzi cation CoM is characterized by big differences and Larsen ' s method

by small differences of values of the bike dif culty. Mamdani ' s method with de-

fuzzi cation CoS is the most reliable.

We choose the bike trail dif

'

culty obtained by Mamdani s method with the

defuzzi cation CoS for another analytical processing. It will permit to reclassify the

attribute of the current bike dif culty of the other roads for

the routing as the finding of the optimal road according to the difficulty.

culty and to add the dif

7.5 Making Use of Mamdani ' s Raster to Finding
the Bike Trail Rating

We assume that Mamdani cation CoS is the main input to our

analysis. The aim of this way is rating of all roads, not only cycle trails, registered
in street net feature class.

' is raster with defuzzi

Table 1 Part of suitable bike trails for membership 0 ; 25 ; 1 h i

Dif culty Small (%) Intermediate (%) Hard (%) All (%)

COS-TM-TM 96.6 83.7 73.7 84.1