Air quality index prediction using IDW geostatistical technique and OLS-based GIS technique in Kuala Lumpur, Malaysia

Abstract

It is known, that the polluted air influences straightforwardly on human wellbeing. Along these lines, the air quality checking surveys the nature of air and recognize defiled territories. Geographic information systems (GIS) provides appropriate tools for the purpose of creating models and describing spatial relationships. This study aims to develop an AQI prediction algorithm based on some meteorological parameters collected using an inverse distance weighted geostatistical technique analysis results, from measurements of three meteorological stations adjacent to the study area Kuala Lumpur of the period June to August 2018. A GIS spatial statistical analysis approach was used. An ordinary least squares (OLS) process was adopted for the 3 months data separately and three models have been obtained. An accuracy value of model performance has been computed were set as (97, 99, and 97%) respectively, specified thru the analysis. So as to test the model, validation applied again using predicted AQI and compared them with observed AQI data, the accuracy was set as (96, 99, and 93%), respectively. The result indicated a very good fit of the OLS model to the observed points, verified that the consequences of these analyses are able to monitor and predict AQI with high accuracy.

Keywords:

• GIS
• AQI
• OLS
• IDW
• meteorological stations

Acknowledgements

The authors would like to thank Dr. Muntadar A. Shareef for his assistance with the statistical programming. The data of this study were obtained from the NASA satellite via RETScreen program funded through the CanmetENERGY Research Center of the Natural Resources Canada’s NRCan. We are also thankful to the RIKEN Centre for Advanced Intelligence Project (AIP), Tokyo, Japan for the funding support.

Disclosure statement

No potential conflict of interest was reported by the authors.

Additional information

Funding

The authors would like to thank the RIKEN Centre for Advanced Intelligence Project (AIP), Japan for the APC funding.