随着我国经济的飞速发展和改革开放政策的不断深入，促进了我国城市化进程加快和城市建设快速发展。随之而来的城市生产安全问题也在不断增多，城市安全生产已成为当今人们十分关注的社会问题之一。城市生产安全事故预测是城市安全生产研究的一项重要内容，它的目的是为了掌握城市安全生产的发展状况，分析存在的问题，有针对性的提出切实可行的管理措施，达到减少城市生产安全事故的目的。 本文从城市生产安全事故预测的紧迫性入手，阐述研究城市生产安全事故预测的意义；对事故的基本理论进行概述；论述国内外常见的事故预测方法，并分别分析其各自的优缺点，针对城市生产安全事故的特点，选出灰色预测法是现在研究城市生产安全事故预测的最佳方法；对邯郸市生产安全事故数据进行统计分析，研究事故发展的规律和趋势；重点介绍灰色预测方法，借助灰色预测在解决既含有已知信息又含不确定信息的灰色系统的优势，以一元一阶灰色模型为基本模型，结合实际进行修正，建立了城市生产安全事故的预测模型；最后，引用邯郸市的生产安全事故数据，应用建立的模型实现了对事故起数、死亡人数等有关参数的预测，结果表明预测精度高，是一种实际可行的预测方法，对预测城市安全形势发展具有重大的现实意义。

Along with the rapid development of economy and social reform in China, the urbanization all over the country has been pushed to a new stage, and the safety issues of industrial productionin cities have become a key problem of social concern. The trend forecast of industrial safety in urban areas can be effectively applied to safety management, analysis of existing problems, and decision making in promoting production safety in an effort to substantially reduce production accidents, thus is very important in urban safety management and control. The urgency and great importance of the urban production safety forcast are elaborated in this article firstly. Then, the normal accident theory and the methods in prediction of accidents at home and broad are overiewed. We compared the differences between the different methods and chose Grey as the best theory in our study. For the characteristics of the urban production safety accidents, selects the grey forecasting is the best method to study now. The statistical analysis of the production safety accident data in Handan were carried on to study the accident safety trends. Considering the advantage of Grey in study of the grey system- with both knwon and unknwon information, we chose GM(1,1) as the fundamental model. Also, we made amends for it according the acctual conditions, and then we established our study model. Basing on this model and the historical data of production accidents in Handan, we predicted the some numbers in the accidents in 2009, such as total number of deaths. The prediction numbers is coincide with the real numbers and with high precision, which indicate that, this model is suitable to the trend forecast of industrial safety in urban areas. And the grey is is one feasible forecasting to predict the development of the safety situation in the city of great practical significance.