Page 219 - Data Science Algorithms in a Week
P. 219

Managing Overcrowding in Healthcare using Fuzzy Logic          203


















                       Figure 8: Fuzzy logic subsystem III.                             Figure 9: Fuzzy logic subsystem IV.


                                        FUZZY LOGIC SYSTEM DEVELOPMENT

                          This section describes the technical process of developing the proposed fuzzy expert
                       system,  which  would  equip  the  designed  framework  with  a  knowledge  base,  a  fuzzy
                       inference  engine,  fuzzifier  and  defuzzifier.  The  knowledge  base  consists  of  a  fuzzy
                       database and a fuzzy rule base, in order to fuel the fuzzifier, defuzzifier, and inference
                       engine portions of the fuzzy subsystems.
                          First, the elicitation of expert knowledge for building the fuzzy database is described.
                       Secondly, this section also describes the process of developing fuzzy rules. Finally, the
                       fuzzification  and  the  defuzzification  processes  are  conceptually  and  mathematically
                       represented.


                       Knowledge Base

                          The knowledge base is an indispensable component of any fuzzy logic system, as it
                       contains both the fuzzy rules base and the database. The development of the knowledge
                       base is keystone for the fuzzy system, and is the most challenging aspect of designing the
                       proposed model. The importance of this knowledge base stems from the dependency of
                       the other component of the system on it, including the fuizzifier, defuzzifier, and fuzzy
                       inference  engine.  Effectively,  the  knowledge  base  is  the  brain  of  the  fuzzy  system,
                       simulating  reasoning  from  a  human  perspective.  The  creation  of  the  knowledge  base
                       involves  systematic  collection  of  qualitative  and  quantitative  data  from  subject  matter
                       experts.  These  experts  have  to  meet  the  following  criteria  in  order  to  be  eligible  to
                       participate in the membership intervals determination and fuzzy rules evaluation:

                            The expert works or has recently worked in Saudi Arabia healthcare institutions
                              for at least five years, or has conducted research in the field of Saudi healthcare.
   214   215   216   217   218   219   220   221   222   223   224