Objective: To understand and predict the situation of occupational diseases in Changsha, and to provide theoretical basis for the scientific formulation of occupational diseases prevention, control strategies and measures. Methods: In April 2019, the data of occupational diseases incidences from 2010 to 2018 were collected. The original GM (1, 1) grey model and buffer operator improved model were established, and compared their prediction effect. The model with the smallest average relative error was selected to predict the incidence of occupational diseases during 2019-2023. Results: The relative accuracy of the original GM (1, 1) grey model and the first-order and second-order buffer operator improved model were 80.92%, 97.71%, 99.64%, respectively. And the c values were 0.74, 0.28, 0.09, and the P values were 0.67, 1.00, 1.00, respectively. It was predicted that the incidence number of occupational diseases in Changsha during 2019-2023 would be 40, 39, 39, 38, 37, respectively. Conclusion: The buffer operator improved model is suitable for the prediction of the original data series with high volatility, and it can fit the incidence of occupational diseases in Changsha.
目的： 了解长沙市职业病发病情况，预测其发病趋势，为制定职业病防控策略与措施提供理论依据。 方法： 于2019年4月，收集长沙市2010-2018年职业病发病数据，建立原始GM（1，1）灰色模型和缓冲算子优化模型，比较其预测效果。选取平均相对误差最小的模型预测2019-2023年长沙市职业病发病情况。 结果： 原始GM（1，1）灰色模型及一阶、二阶缓冲算子优化模型的相对精度分别为80.92%、97.71%、99.64%（c=0.74、0.28、0.09，P=0.67、1.00、1.00）；预测2019-2023年长沙市职业病发病例数分别为40、39、39、38、37例。 结论： 缓冲算子优化模型适用于波动性较大的原始数据序列的预测，可较好地拟合长沙市职业病发病情况。.
Keywords: Buffer; GM (1, 1) grey model; Occupational diseases; Operator; Prediction.