With the economic development, light pollution has become one of the most serious social problems in many countries. In order to measure and reduce light pollution in China, this paper combines light source, economics, environmental and mathematical formulas to develop 6 evaluation metrics and constructs a light pollution evaluation model based on these metrics to measure the light pollution level of a city in China. Specifically, we choose 3 evaluation indicators and 6 metrics, collect data of 31 provinces in China as typical samples, and apply the Entropy Weight Method (EWM) to determine the weight of each metric, using the TOPSIS method to calculate the light pollution risk index (LPRI) of each province. Finally, we classify the 31 provinces into 4 risk levels based on the light pollution risk index by employing the System Clustering method. Moreover, we chose Guangdong and Shanghai as typical locations to validate our model. Comparing our results to the real data, we find that it is feasible and reliable to use light pollution risk index to measure the light pollution. And our results showed that the light pollution evaluation model achieved good performance on the test sample cities.