| Safety accidents is concerned by public, recent years, as increasing attention ofthe country, safety supervision work has improved obviously, but the amount ofaccidents and economic loss is still large,. Analyzing work safety, predict accidentoccurrence, can support the work safety supervision.Work safety and the economic development is closely related, in order to studythe rule of work safety, collecting the work safety monthly fatality data in ShaanxiProvince, establishing the time series, find the clearly seasonal factors, using theX-12-ARIMA method of seasonal adjustment on the industry, mining and businessfatality data of Shaanxi province, get the seasonal factors and long term trend, thestrongest seasonal factor of each year is in July and December, January and Februaryhave the weakest seasonal factor, work safety time series have clearly periodicity.This paper collects different industry fields and cities’ work safety fatalities data,finding that the city which has the better economy gets the worse work safety, Xi’anwhich has the biggest GDP in Shaanxi province gets the largest number of accidentsand work accident deaths, the long-term trend of work accident deaths in ShaanxiProvince decreased spirally; work safety situation of coal industry has improved quiteobviously, but the safety situation of the building industry, fireworks industry, andnon-coal mines have not changed significantly, even getting worse.This paper predicts the future change of work safety situation in Shaanxi Province,establishes the time series model, neural network model and seasonal grey-time seriesmodel,it turns grey-time series model is more accuracy than time series model, in orderto improve the accuracy of prediction, established a portfolio model of three models.The result is consistent with work safety trend of Shaanxi Province,and theaccuracy is reliable,the facalties data of next12months will decrease slightly influctuation,the second half of the year will still be worse than the first. |
PDF Full Text Request