The preliminary criteria of sellectchem environmental protection education in this study obtained from the literature review are community energy-saving solar-assisted heat [24], economical use of natural energy resources [25], resource recovery and reuse, energy-saving materials [26], planning and design of energy-efficient equipment and energy-saving construction [27], urban greening [27], sustainable energy development [23], and energy and environmental efficiency [6].There are totally 16 Delphi experts in this study��four experts of industrial circles, six resident representatives of green community organizations, three experts of governmental authorities, and three scholars with practical experiences.

Since this study aims to promote green community development in Taiwan, the Delphi experts suggested that the environmental protection education in this study should focus on community residents and incorporate more participatory teaching strategies [28]. After an 8-month period of Delphi process in this study, the course contents and teaching methods of environmental protection education beneficial for green community development are confirmed, such as participation of the residents in action performance, watching environmental protection promotion films, enhancing knowledge about energy-saving materials and equipment among the residents, community greening, and problem based learning [29]. The criteria applicable to this evaluation model as shown in Table 1 are obtained with the assistance of the Delphi experts. Each criterion in the table is agreed upon by all the experts.

Table 1Criteria jointly agreed by experts.4.2. Definition of Membership Function, Fuzzy Set, and Fuzzy RangeThe fuzzy logic theory usually uses the Mamdani or Sugeno systems for modeling, where the quantized output values of the Mamdani system are continuous changes, and the quantized output values of the Sugeno system are discrete changes. This study used the Mamdani system for modeling in order to present the continuous output change values of the model. A membership function characterizes a fuzzy linguistic term by giving its support value or degree of membership. The membership value varies from 0 to 1, representing none to full membership. Common membership functions include triangular- and bell-shaped functions [30].

The membership functions in this study are bell-shaped functions for their curves are smoother and, therefore, more suitable for the continuous output changes Cilengitide of the Mamdani system.Table 2 lists the number of fuzzy sets of each criterion: five (Very good, Good, Normal, Poor, and Very poor) in performances activities, three (Good, Normal, and Poor) in energy-saving public facility, three (Good, Normal, and Poor) in energy-saving family; and three (Good, Normal, and Poor) in problem based learning. The fuzzy sets of the four criteria can be combined to form 135 scenarios.