Aplikasi Prediksi Kerentanan Banjir Kota Palangka Raya Menggunakan Metode Fuzzy Mamdani Berbasis Web

Abstract

ABSTRACT

Vulnerability is a condition or condition that can reduce the community's ability to prepare for disaster hazards or threats. The purpose of knowing vulnerability is to reduce the possibility of adverse impacts caused by disasters. The process of predicting flood vulnerability using the Mamdani Fuzzy Inference System method. In the tests carried out by manual calculations and the program uses several defuzzification methods, namely Centroid, SOM (Smallest Of Maximum), LOM (Large Of Maximum), MOM (Mean Of Maximum), Bisector. Furthermore, manual calculations are compared with calculations from the program.

The problem of this research is the process of implementing Mamdani fuzzy logic into a web-based application system to predict flood vulnerability in Palangka Raya City with the PHP programming language.

This research produces an output of flood susceptibility levels, namely safe, vulnerable, very vulnerable, and flood vulnerability. Techniques in data collection that is done by way of observation and literature study. For manual calculations, it begins with determining the fuzzy set of each variable, forming fuzzy rules (implications), the composition of rules using the MAX function, and confirmation (defuzzification). Meanwhile, the program starts with creating an interface design using Balsamiq Wireframes. Then proceed with implementing it into the PHP programming language code so that the design of the flood vulnerability prediction application can function. After the prediction program has been successfully created, then the sub-district monograph data is input into the system for processing with the calculation accuracy results obtained are 100% the same as manual calculations and the accuracy of the prediction results is 80.46%.

 

Keyword: Vulnerability, Fuzzy, Mamdani, Flood, Php.