The year 2019 was the second warmest year in 140 years and the global land and ocean surface temperature experience an anomaly of temperature rise at the average of +0.95 °C, which is only behind 0.04 °C from the record high value recorded in 2016. One of the reasons is the emission of greenhouse gas effects such as carbon monoxide (CO) and nitrogen oxide (NOx). CO and NOx gases are usually being released from automobile vehicles, trucks and from an industrial source such as a gas turbine power plant that uses natural gas/ fossil fuel/ coal as the source of combustion. In order to address this problem seriously, an approach to predict and analyze the CO and NOx emission while optimizing the gas turbine to reach maximum turbine energy yield (TEY) from the gas turbine power plant is being proposed. By applying thermodynamic or statistical methods, a mathematical model can be constructed to predict the emissions using a computer program based on the operating condition of the gas turbine. Low installation, operating and maintenance cost, faster to configure and maintain and also can provide some feasible information for industrial process optimization are the proof why Predictive emission monitoring system (PEMS) is superior compared to Continuous emission monitoring system (CEMS). In this study, ANN-based PEMS for monitoring the emission of CO and NOx gas from the gas turbine power plant will be developed while optimizing the operating condition of the gas turbine for maximum turbine energy yield (TEY). The result shows that the ANN-based PEMS is able to predict CO, NOx and TEY in the range of 0.8 to 0.9 coefficient determination. [ABSTRACT FROM AUTHOR]