Adaptive Conditional Hazard Regression Modeling of Multiple Event Times

Recurrent event time data and more general multiple event time data are commonly analyzed using extensions of Cox regression, or proportional hazards regression, as used with single event time data. These methods treat covariates, either time-invariant or time-varying, as having multiplicative effects while general dependence on time is left un-estimated. An adaptive approach is formulated for analyzing multiple event time data. Conditional hazard rates are modeled in terms of dependence on both time and covariates using fractional polynomials restricted so that the conditional hazard rates are positive-valued and so that excess time probability functions (generalizing survival functions for single event times) are decreasing. Maximum likelihood is used to estimate parameters adjusting for right censored event times. Likelihood cross-validation (LCV) scores are used to compare models. Adaptive searches through alternate conditional hazard rate models are controlled by LCV scores combined with tolerance parameters. These searches identify effective models for the underlying multiple event time data. Conditional hazard regression is demonstrated using data on times between tumor recurrence for bladder cancer patients. Analyses of theory-based models for these data using extensions of Cox regression provide conflicting results on effects to treatment group and the initial number of tumors. On the other hand, fractional polynomial analyses of these theory-based models provide consistent results identifying significant effects to treatment group and initial number of tumors using both model-based and robust empirical tests. Adaptive analyses further identify distinct moderation by group of the effect of tumor order and an additive effect to group after controlling for nonlinear effects to initial number of tumors and tumor order. Results of example analyses indicate that adaptive conditional hazard rate modeling can generate useful insights into multiple event time data.

Adaptive Conditional Hazard Regression Modeling of Multiple Event Times

Recurrent event time data and more general multiple event time data are commonly analyzed using extensions of Cox regression, or proportional hazards regression, as used with single event time data. These methods treat covariates, either time-invariant or time-varying, as having multiplicative effects while general dependence on time is left un-estimated. An adaptive approach is formulated for analyzing multiple event time data. Conditional hazard rates are modeled in terms of dependence on both time and covariates using fractional polynomials restricted so that the conditional hazard rates are positive-valued and so that excess time probability functions (generalizing survival functions for single event times) are decreasing. Maximum likelihood is used to estimate parameters adjusting for right censored event times. Likelihood cross-validation (LCV) scores are used to compare models. Adaptive searches through alternate conditional hazard rate models are controlled by LCV scores combined with tolerance parameters. These searches identify effective models for the underlying multiple event time data. Conditional hazard regression is demonstrated using data on times between tumor recurrence for bladder cancer patients. Analyses of theory-based models for these data using extensions of Cox regression provide conflicting results on effects to treatment group and the initial number of tumors. On the other hand, fractional polynomial analyses of these theory-based models provide consistent results identifying significant effects to treatment group and initial number of tumors using both model-based and robust empirical tests. Adaptive analyses further identify distinct moderation by group of the effect of tumor order and an additive effect to group after controlling for nonlinear effects to initial number of tumors and tumor order. Results of example analyses indicate that adaptive conditional hazard rate modeling can generate useful insights into multiple event time data.

高校师德师风舆情事件的特征、趋势风险与对策建议

近年来,高等教育领域零星出现一些师德师风失范行为的网络舆情,不仅为高校管理带来了巨大压力,也对高校的稳定发展和高校形象带来负面影响。本文首先对近年来高校师德师风舆情事件进行系统、客观的梳理分析,明确当前高校师德师风舆情事件的类型、传播及群体特征。进而结合媒介传播尤其是新媒体传播特点,从失范行为、涉事主体、高校应对诸方面归纳高校师德师风舆情事件的趋势与风险。在此基础上,提出针对性的解决对策与建议,以期对高校相关部门客观认识、理性应对和科学防范师德师风舆情事件有所助益。研究发现,近年来师德师风舆情事件主要涉及高校教师言行不当、学术不端、违反工作纪律等方面。此类舆情事件往往借助自媒体传播,呈现出突发性强、传播速度快、传播面广、难以控制的特点,处置难度较大,容易成为舆情爆点,并引发民众对教师的信任危机。部分高校舆情风险意识不强,舆情应对能力欠佳,沟通机制缺乏。基于此,高校应高度重视涉教师师德师风舆情工作,提升舆情风险意识;完善师德师风建设机制,拓宽师生表达空间;强化师德师风教育,做好正面宣传工作;及时回应公众诉求,维护网络空间秩序。

基于事件相关电位的监控员任务切换刺激模态差异性对警觉度的影响研究

为探究不同模态转换下任务切换对监控人员警觉度衰减现象的影响,采用脑电实验法,设计了危险动作识别、危险场景识别等任务,并结合行为数据及事件相关电位(Event Related Potential,ERP)技术进行综合分析;实验重点考察在任务切换过程中,不同模态的任务之间进行切换对监控人员警觉度的具体影响;通过对煤矿监控调度人员的实证研究,揭示了切换任务的模态差异性对监控员警觉度的影响及其内在机制。结果表明:当在相同刺激任务间进行切换时带来的工作绩效下降,主要原因在于任务切换产生的切换成本导致注意力分散,具体体现在任务切换后P300和P200平均波幅的升高;但是在不同刺激任务间进行切换时,尤其是切换到听觉刺激任务时,被试者的警觉度水平反而得到了一定程度的提升。

湖泊沉积的灾害事件记录与识别方法:研究进展与存在问题

湖泊沉积储存有丰富的信息,常被用于指示气候变化。地震、洪水、滑坡或崩塌等自然灾害事件通常会在湖泊沉积中形成特征性的湖泊沉积扰动,可以使用湖泊地质学方法对其进行分析与识别。本文回顾了湖泊沉积研究历程,对典型自然灾害湖泊沉积特征进行系统总结,着重梳理介绍湖泊沉积中自然灾害事件识别方法。其中综合分析法集成了现阶段湖泊事件沉积识别的多种分析手段:(1)根据沉积相与异常事件沉积(EID)之间的关系判定其内源或外源成因;(2)根据对触发因素的排除,确定最可能的触发因素;(3)根据现有的识别标准,验证最可能触发因素的正确性。是目前识别湖泊扰动成因最有效的分析手段。本文还对湖泊沉积灾害事件研究的优势与不足及制约因素进行讨论,湖泊沉积具有连续性、敏感性和高分辨率的特点有助于对灾害事件进行研究,但对湖泊典型自然灾害事件沉积识别标志的建立仍然是制约湖泊异常事件沉积研究的关键。

基于深度学习与深度估计的施工机械危险区域侵入智能预警方法

为解决因工人和施工机械侵入施工危险区域等原因造成的工程安全事故问题,提出一种多任务驱动的施工机械危险区域侵入事件动态识别与预警方法。首先以置换可变性卷积DConv2模块的Yolov8网络进行目标类别检测和坐标外轮廓提取,提高移动施工机械的识别准确率;然后结合Monodepth2单目深度估计网络进行深度信息估计和坐标统一,计算工人或施工机械距离危险区域事件的实际距离,用于评估和预警危险区域侵入风险;最后将测试结果与不同修改层的Yolov8、原始Yolov8和Yolov5模型进行对比,并设计4种场景进行模型性能验证。结果表明:模型在施工机械的识别和轮廓提取精度上分别提高了2.99%和3.55%,对工人和施工机械侵入移动施工机械危险区域风险事件的识别准确率能保持在88%以上,FPS保持在17.7左右,可以有效实现对施工机械危险区域侵入事件的智能动态预警。

Risk Assessment of an Environmental Liability in the Commercialization of Hydrocarbons-A Secondary Publication

The risk analysis evaluates the probability of occurrence with the impact of said risks.In the hydrocarbon sector,an environmental liability is an improperly abandoned facility,soil contaminated by spills or others.The objective of this work is to define a risk management plan.The characterization of the residue and the area was carried out.Hazardous events were identified:overflow spill,soil contamination by infiltration,as well as radiant heat and suffocating and toxic fumes from a fire.The matrix method was applied considering the probability of occurrence against vulnerability.The danger of infiltration implied a high level of risk.The risk management plan was drawn up,its main action being the liquidation of the pool and the sanitation of the area.

环境事件中不明来源固体废物危险特性鉴别——以四川省某地仓库贮存物为例

以四川省某地仓库内非法贮存不明固体废物事件为例,通过对不明固体废物的性状、物相组成、成分分析进行研究,判断该不明固体废物的产生来源。根据其来源判断其可能具有的危险特性,进一步开展实验检测明确其危险特性。经鉴别,该材料为二次铝灰,是列入《国家危险废物名录(2021年版)》中的危险废物,类别为HW48有色金属采选和冶炼废物,并进一步明确了二次铝灰的危险特性为遇水释放氨气的反应性和氟化物的毒性。

膏状烟气处理渣危险废物鉴别实例分析

本研究以某环境事件中非法倾倒的膏状固体废物为研究对象,通过成分、浸出毒性、腐蚀性pH值检测确定固体废物来源与收集的资料信息相吻合。通过对照《危险废物鉴别标准(征求意见稿)编制说明》确定浸出毒性铊指标参考限值,根据样品浸出毒性铊指标检出结果判定该固体废物为HW30类危险废物,废物代码为“900-000-30”。本研究通过对环境事件实操案例的分析,指出在现有鉴别标准无限值时,应综合考虑潜在危险特性对生态环境和人体健康的损害,依照编制说明中提及的原理设定指标参考限值,旨在为危险废物鉴别单位和技术人员提供可参考的技术工作方案。

生态环境事件中固体废物危险特性应急鉴别技术研究

企业生产产生的固体废物种类多样、成分复杂,若未经妥善处理处置,将会对生态环境造成不可逆转损害,同时其中具有危险特性的危险废物进入环境后损害更为严重。以某环境事件中的污泥样品为研究对象,根据待鉴别对象理化性状情况,采集代表性样品5个,采用先有机定性扫描后进行浸出毒性苯、甲苯、二甲苯指标定量检测的鉴别技术路线,最后根据样品检测结果判定待鉴别对象为具有甲苯浸出毒性危险特性的危险废物。本研究采用“定性-定量”相结合技术方法进行固体废物环境应急鉴别工作,具有样品采集数量少、指标检测具有针对性、成本低、鉴别时间短等优点,值得在环境事件固体废物应急鉴别工作中推广应用。

Comparing rockfall hazard and risk assessment procedures along roads for different planning purposes

Hazard and risk assessment procedures of different types of rockfall were analyzed to compare their outcomes when they are applied to the same case study.Although numerous methodologies are available in literature,rockfall hazard and risk analyses are often limited to standard estimations,affected by a margin of uncertainty,especially when relevant engineering projects are about to be realized.Based on the design purpose,different types of approaches can be chosen among the qualitative and quantitative ones available in literature,which allow different levels of analysis.One of the main criticisms related to rockfall events is the risk affecting linear structures,such as road or railways,due both to their strategic relevance for trade and communications and to the great entity of the exposed value(traffic units)traveling along them.In this perspective,a comparison between the qualitative method known as Evolving Rockfall Hazard Assessment(EHRA),the semi-quantitative modified Rockfall Hazard Rating System(RHRS)and the quantitative Rockfall Risk Management(RoMa)approach is herein commented according to a practical application to a case study.It is the case of the rockfall threat along slopes crossed by a strategic road connecting two of the most known spots of eastern Sicily(Italy),at the Taormina tourist complex.Data were retrieved from both recent literature and technical surveys on field.Achieved results highlight how the approaches are affected by a different level of detail and uncertainty,arising also by some necessary assumption that must be taken into account,especially when mitigation measures or territory planning have to be designed.Achieved results can be also taken into account for similar studies worldwide,in order to choose the most suitable procedure based on the design purpose.This is indeed crucial in the perspective of the optimization of time and economic resources in the territorial planning practice.

关于路桥跨越敏感水体时事故应急池的设置探讨

危化品运输车在跨越敏感水体路桥上行驶时,若发生交通事故将会导致附近水体污染,所以需要设置事故应急池。但现行跨越敏感水体的事故应急池设置既无明确规范,容积计算又存在分歧。因此,为明确其设置要求和计算方法,通过分析事故应急池容积的影响因素,总结了其计算方法;明确提出了路桥跨越敏感水体且有危化品运输事故风险时,应设置事故应急池;最后对事故应急池全过程管理和预警等方面问题,给出了相应的策略和建议。

基于有限状态机的预期功能安全危害识别方法

针对自动驾驶系统危害与场景不可分割的特点,提出了一种基于有限状态机模型(FSM)的整车级预期功能安全危害识别方法。首先,明确危害事件组成要素;其次,将自动驾驶系统抽象为有限状态机模型以明确车辆状态和运行环境;最后,通过识别车辆状态与运行环境的冲突情况,系统性识别自动驾驶系统预期功能安全危害事件,减少对专家知识的依赖。为验证所提出方法的有效性,在某SAE L3级自动驾驶汽车上应用了该方法进行危害识别。结果表明,相较于系统理论过程分析(STPA)方法,有限状态机模型包含更加详细且系统化的环境信息,且由有限状态机模型直接输出危害事件要素,提高了危害识别的系统性。

Spatial Analysis of Affected Areas by Extreme Hydrological Events in Rio de Janeiro (The Host City for the 2016 Olympic Games)—Brazil

The most frequent types of disasters in Brazil are associated with extreme hydrological events. Adding to this situation, in Brazil, there are the extreme sociospatial discrepancies that are historically constructed. They leave marks in space, making specific areas, for political interest or not, more vulnerable to the different mentioned processes. To this extent, the public management should plan actions and, also, it may act in a more holistic way adding new trends, such as, the use of geotechnologies that are applied to environmental management in the realm of urban planning. From a case study in the city of Rio de Janeiro, this paper aims to contribute and draw attention to such issues. It may reach its target through pointing and spatially analyzing which areas are in the most critical situation related to the occurrence of extreme hydrological events. To achieve this goal, it was proposed a methodology to inventory the occurrence of inundation/flooding for a certain period, in this case 2001-2008. Besides, the methodology could systematize data and integrate them with other important issues for supporting the process of identification and analysis of the most critical areas. Historical aspects of occupation and the most vulnerable socio-environmental aspects were raised, too, in order to validate the highlighted areas. This integration was enabled concerning the support of geoprocessing techniques. The results of this integration subsidized the mapping and spatial analysis of the affected areas in the city and the criticality in relation to the extreme hydrological events. Finally, it was possible to observe that the identification of the most critical affected areas does not mean the exhaustion of the subject. The location of these areas is directly linked to the sample that was used as the basis for analyses. This data set only represents recorded events by the Municipal Civil Defense respecting the studied time frame. However, the results permit to take notes and raise relevant environmental questions about the roots and the occurred impacts that were originated from the extreme hydrological events, which have been observed in recent decades.

基于危险预知训练模式的护理不良事件风险管理成效

目的探讨危险预知训练模式在护理不良事件风险管理中的应用效果。方法采取前瞻性类实验研究设计,2022年1-6月收治患者13346例纳入干预前组,2022年7-12月收治患者11520例纳入干预后组。干预前组采用常规护理不良事件管理方法,干预后组采用基于危险预知训练模式的护理不良事件风险管理。比较干预前后护士安全行为得分,以及2组患者不良事件、投诉纠纷发生率。结果干预后护士安全行为得分较干预前显著提高(P<0.05);干预后组患者用药错误、跌倒、非计划性拔管类不良事件、投诉纠纷发生率显著低于干预前组(均P<0.05)。结论基于危险预知训练模式的护理不良事件风险管理能够有效提高护士的安全行为,降低不良事件、投诉纠纷的发生率。

The Additive-multiplicative Hazards Model for Multiple Type of Recurrent Gap Times

Recurrent event gap times data frequently arise in biomedical studies and often more than one type of event is of interest. To evaluate the effects of covariates on the marginal recurrent event hazards functions, there exist two types of hazards models： the multiplicative hazards model and the additive hazards model. In the paper, we propose a more flexible additive-multiplicative hazards model for multiple type of recurrent gap times data, wherein some covariates are assumed to be additive while others are multiplicative. An estimating equation approach is presented to estimate the regression parameters. We establish asymptotic properties of the proposed estimators.

基于Copula函数的绕阳河流域降水事件危险性评估

利用1961—2021年绕阳河流域周边8个气象站的逐日降水资料,以降水持续时间和降水强度作为判断降水事件的两个要素,运用Copula函数确定降水事件的联合概率和重现期,并构建危险性评估模型,分析绕阳河流域降水事件危险性空间特征。结果表明:绕阳河流域周边不同地区降水事件的持续时间和降水强度的最大值存在显著差异。Lognormal和Logistic函数对两个降水要素的拟合效果较好,Clayton函数适合反映两要素联合下降水发生的概率。随着降水持续时间和降水强度的增加,极端降水事件的危险性快速升高;不同地区降水事件的概率和重现期存在着显著差异,阜新站降水持续7 d,强度达到60 mm的事件每10 a出现一次,而盘锦站和北镇站发生持续6 d以上的降水事件较少,多为20 a一遇;降水事件危险性高值区处于绕阳河流域南部,主要位于盘锦、辽中和北镇。

暴雨过程致灾危险性评估方法研究——以孝感市“8·12”暴雨过程为例

暴雨过程致灾危险性评估是暴雨灾害风险评估的基础和重要环节,可为气象防灾减灾提供重要技术支撑。利用孝感市7个国家气象观测站及120个区域气象观测站逐日、逐时降水量观测资料及地理信息资料,采用暴雨过程灾害危险性动态评估方法,对2021年8月12日发生于湖北省孝感市的一次区域性暴雨过程进行了致灾危险性评估。基于暴雨过程降水特征量分析,确定暴雨过程致灾因子,采用信息熵权法确定4个致灾因子权重系数,计算了暴雨过程强度指数;基于暴雨区下垫面特征分析,明确地形影响系数、水系影响系数为孕灾环境因子,并等权求和构成孕灾环境综合指数;暴雨过程致灾危险性指数由暴雨过程强度指数和孕灾环境综合指数加权构成,采用自然断点法将暴雨过程致灾危险性指数划分为5个等级,基于GIS绘制了暴雨过程致灾危险等级分区图,利用主要承灾体灾情数据验证了危险性评估结果的合理性。结果表明:(1)此次暴雨过程具有来势猛、短时降水量大、降水集中等特点,暴雨过程86.1%降水量集中在6 h以内,最大6 h降水量达50 a一遇,选取致灾因子时重点考虑短历时降水量;(2)孝南区大部、云梦县中部、应城市东北部、安陆市西南部为暴雨过程致灾高危险区,而较高危险区覆盖孝感市中部,这与暴雨强度及地形、水系河网等环境敏感性有关;(3)高和较高危险区对应暴雨灾情较重的区域,实际灾害等级达到“大型”灾害等级,致灾危险性评估结果反映了主要承灾体灾害损失实际。

Kaiser模型在医院院科两级灾害脆弱性分析中的应用

目的 灾害脆弱性分析是医院应急管理中的重要内容,Kaiser模型作为医疗机构灾害脆弱性分析的重要工具,在使用过程中却缺少相应的标准和规范,本研究旨在通过Kaiser模型在医院和科室灾害脆弱性分析中的使用,探索如何更加规范地使用Kaiser模型。方法 2021年3月分别由选取的相关人员对某医院和医院某临床科室地潜在的风险事件进行筛选,结合Kaiser模型设计相应调查表进行调查,并录入Excel 2007进行统计分析,潜在风险事件根据Kaiser模型评分表的最后评分进行排序,以明确医院和科室当年潜在风险事件的应对重点。结果 灾害脆弱性分析较客观地反映了医院和科室对突发事件的承受能力及目前需要优先应对的突发风险事件,对医院的应急管理具有重要作用。结论 Kaiser模型是医院应急管理的重要工具,如何更好地使用这个工具,使其能真正地反映客观实际,需要在使用过程中更多地探索标准化规范化使用过程。

生态环境事件中固体废物溯源鉴定与应急处置技术研究

基于工业固体废物鉴定内容和常用技术路线,结合生态环境事件中固体废物鉴定特点与难点,总结生态环境事件中固体废物鉴定与溯源技术要点,提出应急处置技术要点。指出生态环境事件固体废物鉴定与溯源中应重视现场踏勘重要性,充分发挥现场快速检测手段作用,做好样品采集与记录工作,落实现场安全防范工作等,应急处置中应做到立即启动现场控制和尽快明确固体废物属性的有机结合,精准、高效地做好生态环境事件固体废物溯源鉴定与应急处置工作。