基于灰色-马尔科夫链理论的建筑施工事故预测研究

Research on Construction Accident Forecast Based on Gray-Markov Theory

我国建筑事故预测中具有统计数据少、数据波动性大等特点,缺乏具有高精度、可操作性强的预测模型。将灰色系统理论和马尔可夫原理相结合,充分发挥了灰色系统理论适用于小样本数据拟合和马尔可夫适合处理数据波动大的系统过程的优势,通过GM(1,1)模型的建立,提出一种适合建筑事故统计数据特点的灰色马尔可夫预测方法。将该方法应用于1994—2007年建筑施工事故次数分析,以此为基础对2008—2009年的建筑施工事故次数进行预测。研究结果表明:基于灰色马尔可夫理论建筑事故预测结果精度可达90%以上。

The construction accident forecast statistical data in China have the characteristics of poverty and great fluctuation, and the existing forecast model has a lack of accuracy and operability. For this, Markov and Gray theory was combined by GM （1,1） to build the Grey-Markov forecasting model that adapts to the characteristics of construction statistical data. This model gives full play to the superiority of Grey theory in fitting small samples and the advantages of Markov theory in dealing with fluctuation data. Based on the application of this model in the number analysis of construction accidents from 1994 to 2007, it was further used to forecast the number of construction accidents in 2008 and 2009. The results show that the accuracy of accident forecast method based on Grey-Markov theory is up to 90%.