Page 229 - Data Science Algorithms in a Week

Managing Overcrowding in Healthcare using Fuzzy Logic 213

Table 6: Parameters of fuzzy subsystem I’s, II’s, III’s, and IV’s membership functions.

The trapezoidal membership functions in Figure 19 represent the four classes used
for the boarding input in subsystem IV. Boarding was considered to be “very high”
between values of 0.26 and 0.4, making its core structure the largest while indicating the
largest interval of values where a class was assigned full membership. Between the
values of 0.16 and 0.32, boarding was considered “high”, which is associated with the
smallest membership function core structure belonging to the high class. The low and
medium classes existed between the intervals of [0, 0.12], and [0.04, 0.24], respectively.
Crowding, the final output of the system, is represented by the triangular membership
functions in Figure 20. The linguistic terms “insignificant”, “low”, “medium”, “high”,
and “extreme” were associated with the five classes. The membership functions were