Implementation of intuitionistic fuzzy inference systems to assess air quality forecast: Case of Malaysia

Artificial intelligence models have been widely applied in real-life problems due to its flexibility to adapt the existence of crucial conflicts within the problems. Prompt development of alternative methods such as the data-driven models cover major drawback of conventional methods regarding the issues of uncertainty, vague and inadequacy of information. Here, we discuss the application of intuitionistic fuzzy set inference systems; an extensions of fuzzy theory. The applicability of the model to forecast atmospheric data is validated by data of PM10 concentration in Malaysia. Results show an encouraging predictability justified by the statistical errors RMSE, MAE and MAPE and compared to neural network model. The paper discusses some of its theoretical and the situation of air pollution (particulate matter) which could facilitate the design and management of potential detrimental impacts caused by the driving parameters. Although the current study is intended for Malaysian air quality, the findings can be generalized for other countries with similar driving forces of the air pollution.