The Application of Fuzzy Interval Correlation Evaluating the Relationship Between Transportation Engineering and Air Pollution - Innovative Management in Information and Production - page 271

Tabl e 2 Correlations interval based on passenger counts and the air pollution in Taiwan

Fuzzy correlation TSP ASP SO 2 O 3 Fallout

By definition 1 (-.178, -.142) (.325, .420) (.356, .379) (.370, .425) (-.181, -.153)

By definition 2 (-.187, -.073) (.273, .335) (.166, .552) (.285, .437) (-.163, -.150)

In Table 2 , we have the following findings. First, besides the correlation of
passenger counts, the TSP and fallout are low significance negative by schemes
of definition 1 and definition 2, and this result denotes that the passenger counts
of Taipei MRT system increase; then that can reduce the value of TSP and fallout.
Second, the correlation coefficient is middle level for passenger counts and the ASP,
SO 2 ,and O 3 by the approach of definition 1; this means the values of ASP, SO 2 ,and
O 3 have a lot of effect to the passenger counts. Third, the correlation coefficient is
of low significance for passenger counts and the ASP, SO 2 ,and O 3 by the approach
of definition 2; this means the values of ASP, SO 2 ,and O 3 have a little effect to
the passenger counts; this result shows that the passenger counts will affect the air
pollution, such as the air pollution of ASP, SO 2 ,and O 3 , can be affected by the
passenger counts of Taipei MRT system.

5 Conclusions

In the progress of the scientific research and analysis, the uncertainty in the
statistical numerical data is the important point of the problem where the traditional
mathematical computation is hard to be established. If we achieve this artificial
accuracy to do causal analysis or measurement, it may lead to the deviation of
the causal judgment, the misleading of the decision strategy, or the exaggerated
difference between the predicted result and the actual data. As the pattern of
data of interval occurred in transportation engineering or energy environment, our
proposed methods can be applied to make management strategy and decision as
the two variables that illustrate this kind of fuzzy interval data. In other words,
this paper employs a simple approach to derive from fuzzy interval measures based
on the traditional definition of Pearson correlation coefficient which are easy and
straightforward. In the formula we provided, when all observations are real numbers,
the developed model becomes the classical Pearson correlation formula. In practice,
many applications are fuzzy in nature. We can absolutely ignore the fuzziness
and make the existing methodology for crisp values. However, this will make the
researcher overconfident with their results. With the methodology developed in this
paper, a more realistic correlation is obtained, which provides the decision maker
with more knowledge and confidence to make better strategies.