Thesis Archive
A Rule Based Decision Model for Prioritizing Renewable Energy Systems
Big-asan, Ann Dannielle
Lee, Angela Mae
Abstract:
It is predicted that energy consumption in the Philippines would increase in the future. With the Philippines’ shift towards sustainable renewable energy (RE) systems according to government policies, it is crucial that the country is able to discern which RE it should prioritize given the Philippine setting. Normally, this selection and prioritization is done by decision makers and experts, however, in their absence, the issue becomes a Multi-Criteria Decision Analysis (MCDA) problem. To avoid the cumbersome procedures of evaluating alternatives and criteria brought by deductive MCDA techniques, an inductive MCDA technique is explored through the application of the rough set theory (RST). RST makes decisions based on learning from previous examples. As such, the required inputs are the previous rankings of expert/s that act as decision attributes, the performance parameters that act as conditional attributes, and the alternatives being compared themselves. Qualifications of experts include but are not limited to: having at least a bachelor’s degree in the related field, having their licensures, and having worked in the field for at least 1 year. This information is then constructed into a pairwise comparison table that acts as the decision table to be discerned through the rough set analysis. From the pairwise comparison table, rough set rules are generated which, after careful analysis and evaluation of the generated rules, can be used to rank new alternatives that may be developed in the future. The aforementioned methodology for developing a rule-based decision model is performed in two case studies, where the 1st case acts as a basis for the procedure used in the 2nd case. While the rough set analysis illustrates similarity for the rankings obtained from a Fuzzy Analytic Hierarchy Process (FAHP) approach in the 1st case, the rankings it generated from the 2nd case differ from the rankings of the individual experts surveyed. This may be due to the fact that there is no unity or clarity among the priorities placed by experts on the parameters that can be used by the rough set to classify decisions. Moreover, the difference in ranking can also be attributed to the fact that there may also be other factors that were subconsciously considered by the experts but were not included in the decision table. Nevertheless, the study successfully fulfilled its objectives and may be used as basis for further studies in this field of research.
Adviser:
Aviso, Kathleen