In order to solve the problem that there are many indoor environment factors and the interaction relationship is complex, which affects the precise decision-making of indoor environment comfort, an indoor environment control decision-making system based on improved D-S evidence theory is designed. Firstly, the boxplot method and the mean substitution method are used to detect and repair the abnormal collected data. Secondly, the distance adaptive weighted fusion algorithm is used to realize the first level fusion of similar sensor data. Finally, the improved D-S evidence theory algorithm is adopted to realize the global fusion decision. The experimental results show that the improved D-S evidence theory algorithm can effectively avoid the fusion decision error caused by the conflict evidence. The system can realize the precise decision of indoor environment control and the fusion decision accuracy is high, which has a certain promotion and application value.