Power producers are looking for ways not only to improve the efficiency of power plant assets but also to alleviate concerns about the environmental impact of power generation without compromising their market competitiveness. To meet this challenge, advanced analytics through data mining approaches has been applied to predict and explain the leading causes of variation in heat rates and the opacity of the flue gas exhaust emissions, which are key factors in measuring the overall efficiency of the power plant. The results from a case study from a coal-fired power plant in Thailand demonstrate the practical validity of our approaches. The key finding contributes to identifying parameters that are the key determinants of excessive heat rates and opacity levels. Thus, corrective and preventive actions related to those parameters can be regularly evaluated and monitored. (C) 2014 Elsevier Ltd. All rights reserved.