The Amount of Solid Waste Forecasting using Time Series ANFIS

Abstract

Due to the increasing of population in Thailand and not well educated on what recycling is and how an individual’s actions can make a difference. It caused problems related to solid waste management in Bangkok. This paper aims to develop a predictive model to forecast the amount of solid waste using Adaptive Neuro-Fuzzy Inference System (ANFIS). The solid waste data was collected from Bangkok in total of fifty districts between October 2002 and December 2015 to support the decision system for solid waste management and a solid waste controlling promotion guideline. The data was collected with no missing values. Therefore, the filtering is consided with an outlier that it is significantly different from the group or divergent from the other data values. The Z-score is used to measure a score's relationship with the mean value in a group of scores. An ANFIS model and data analysis have been investigated and performed using MATLAB. The performance result is given quite good values in terms of Root Mean Square Error (RMSE) and Mean Absolute Error (MAE).