Sakarya University Journal of Science

Öz

The most important reality in the business
life that has not changed in the last fifty years is perhaps the necessity for
the “customer satisfaction to be sustainable.” Every failure that adversely
affects product quality also causes customer dissatisfaction. In this study,
the Failure Mode and Effects Analysis (FMEA) was used to analyze the potential
quality failures of the production system in a textile factory. By using this
method, the probability, severity and detectability of quality faults (quality
risks) which could lead to customer dissatisfaction were determined. In this
method, the risk magnitudes are found by multiplying the probability, severity
and detectability values of risks. These risks with high priority, which are
also called the Risk Priority Numbers (RPN), are the risks which need to be
considered as priority, and for which more resources are needed to be
allocated. These three components are equally effective when determining the
Risk Priority Number because of the multiplication operation. However, when
ranking the risks, the role of the severity component is more important than
the others. This is because a risk of low severity may rank low in the priority
order even if it occurs very frequently (even if it has a high probability).
Similarly, in the exact opposite condition, even if the probability is low, a
risk with a high severity needs to be placed higher in the priority order, and
more resources are needed to eliminate such risks. Due to this uncertain
situation, prioritization has also been made by creating a rule-based fuzzy
logic in MATLAB, with the assumption that it is more meaningful to use fuzzy
expressions instead of definite expressions when determining the magnitudes of
risks.