Page 213 - Data Science Algorithms in a Week

Managing Overcrowding in Healthcare using Fuzzy Logic 197

stakeholder’s perspectives and has the ability to be applied in variety of healthcare
systems.

FRAMEWORK DEVELOPMENT

As shown in Figure 1, the proposed framework encompasses four components,
including the crisp inputs, a fuzzy logic system, the expert knowledge, and crisp outputs.
The figure further illustrates the relation between these components. While a fuzzy
system alone may be simple to design in general, what makes this framework novel is its
integration of expert knowledge in the form of a knowledge base with the fuzzy system.
The crisp inputs include identified measures and indicators that reflect many ED and
hospital operational aspects that affect ED’s crowding levels. The crisp inputs feed the
second component of the framework, the fuzzy logic system, with numerical information.
The fuzzy logic system includes the fuzzifier, fuzzy inference engine, knowledge base,
and defuzzifier, at which the crisp ED crowding measures are converted to crisp output.
Expert knowledge is used to construct knowledge base, consisting of the fuzzy rules and
the database, which fuzzifies inputs, provides supporting decision making information to
the inference engine, and defuzzifies outputs. The resulting crisp output reflects the level
of overcrowding in the ED. The output of the framework is an index of ED overcrowding
that aids in measuring patient congestion and patient flow within EDs. It is a quantitative
instrument that evaluates the ED crowdedness based on the input of healthcare experts.
The output can be utilized with a decision support system to inform and aid an ED in
coping with ED crowding.

Figure 1: Proposed framework.