Formulation of cleanup standards for trace elements with probability plots

Probability plots of trace element concentrations may be used to partition, or segregate, sample data into its constituent populations. Two populations are typically present in data from sites requiring remedial action, one representing clean, uncontaminated soil or sediment (background) and the other representing contaminated ground. The use of such plots in the analysis and evaluation of environmental data permits a statistical characterization of the background populations, from which defensible cleanup criteria may be developed. These criteria will be environmentally conservative, yet will minimize the amount of soil removed in a remedial action. An example is given from a successful cleanup of a surface impoundment.     

事故损失统计中的量化问题

简要介绍了事故损失统计中一些量化方法的应用,同时提出用方差分析来确定加权系数的取舍。     

灾害损失评估及其系统结构

从统计学的角度对灾害损失评估的方法、内容及其系统结构进行了简要介绍,可以为灾害损失评估系统及灾害统计学学科的建立提供参考。     

煤矿安全系统的计算机统计控制方法

煤矿安全系统在稳定状态，安全监测设备的数据符合正态分布。无事故发生时，控制图表示监测数据样本均值和极差的数据应以９９．７％的概率落在控制区内。运用统计学理论和方法分析、控制均值和极差点的位置及点间关系，可以达到预测和控制安全系统事故的发生和发展的目的。     

A study of maintenance-related major accident cases in the 21st century

This paper is based on a review of 183 detailed, major accident investigation and analysis reports related to the handling, processing and storage of hydrocarbons and hazardous chemicals over a decade from 2000 to 2011. The reports cover technical, human and organizational factors. In this paper, the Work and Accident Process (WAP) classification scheme is applied to the accident reports with the intention of investigating to what extent maintenance has been a cause of major accidents and what maintenance-related causes have been the most frequent.The main objectives are: (1) to present more current overall statistics of maintenance-related major accidents, (2) to investigate the trend of maintenance-related major accidents over time, and (3) to investigate which maintenance-related major accident causes are the most frequent, requiring the most attention in the drive for improvement.The paper presents statistical analysis and interpretation of maintenance-related major accidents’ moving averages as well as data related to the types of facility, hazardous substances, major accidents and causes. This is based on a thorough review of accident investigation reports.It is found that out of 183 major accidents in the US and Europe, maintenance was linked to 80 (44%) and that the accident trend is decreasing. The results also show that “lack of barrier maintenance” (50%), “deficient design, organization and resource management” (85%) and “deficient planning/scheduling/fault diagnosis” (69%) are the most frequent causes in terms of the active accident process, the latent accident process and the work process respectively.     

一种基于小样本数据信息扩散的重大火灾频度估算方法

重大火灾的发生是小概率事件,其历史数据属于小样本统计数据。在应用频率直方图法对小样本数据进行处理时,分析结果对直方图的起点、区间步长以及样本中的奇异数据十分敏感。采用信息扩散方法,对日本1995-2008年各年重大火灾次数统计样本进行分析,对比了不同扩散条件下的扩散结果,结果表明信息扩散方法具有很好的稳定性和一致性。基于信息扩散的结果,计算了以年为周期的重大火灾发生次数的超越概率分布,建立了一种重大火灾频度估算方法。     

环境空气质量统计预报模式的研究——沈阳市环境空气质量各季节预报模式

利用沈阳市大气自动监测数据和同期气象资料,采用统计预报方法开发研制了环境空气质量预报模式,预报模式包括春、夏、秋、冬四季预报方程。预报对象为:可吸入颗粒物(PM10)、二氧化硫(SO2)、二氧化氮(NO2)日均值,经过近5年的使用,预报效果良好。     

基于CHNS的中国城乡居民的食品消费特征——兼与国家统计局数据对比

基于2011年我国12个省（直辖市）的中国健康与营养调查数据（China Health and Nutrition Survey, CHNS）,文章利用食物成分表重新估算我国城乡居民主要食品的消费量及在外消费特征,并与国家统计局公布的官方数据进行对比。结果显示：1）口粮方面,2011年我国居民平均消费口粮量为119.93 kg,外出消费比例为9.91%。2）肉蛋奶类方面,2011年我国人均肉类消费量（猪牛羊禽）为32.84 kg,其中,猪肉消费量占69.85%,禽肉消费量占20.98%,牛羊肉消费量偏低,为9.20%。我国居民平均还消费蛋类10.90 kg、奶类13.56 kg。3）我国城镇居民口粮消费明显低于农村居民,肉蛋奶消费量明显高于农村居民,但农村居民肉蛋奶在外消费的比例要高于城镇居民。4）国家统计局公布的肉蛋奶数据整体上偏低,农村消费数据的偏低程度尤为严重,重点体现在鲜蛋、水产品与鲜奶产品上。国家统计局公布的食品消费数据多与CHNS数据的家庭内部消费数据相吻合,其数据偏小的原因可能是由于缺失户外消费数据的统计。从消费数量上来看,预计今后我国居民口粮消费量的下降空间已经非常有限,但肉蛋奶消费水平仍有较大的抬升空间,农业的生产压力与畜牧饲料大量进口的局面将持续存在。     

Fabricating data: how substituting values for nondetects can ruin results, and what can be done about it

The most commonly used method in environmental chemistry to deal with values below detection limits is to substitute a fraction of the detection limit for each nondetect. Two decades of research has shown that this fabrication of values produces poor estimates of statistics, and commonly obscures patterns and trends in the data. Papers using substitution may conclude that significant differences, correlations, and regression relationships do not exist, when in fact they do. The reverse may also be true. Fortunately, good alternative methods for dealing with nondetects already exist, and are summarized here with references to original sources. Substituting values for nondetects should be used rarely, and should generally be considered unacceptable in scientific research. There are better ways.