Observer Error When Measuring Safety-Related Behavior: Momentary Time Sampling Versus Whole-Interval Recording

Interval recording procedures are used by persons who collect data through observation to estimate the cumulative occurrence and nonoccurrence of behavior/events. Although interval recording procedures can increase the efficiency of observational data collection, they can also induce error from the observer. In the present study, 50 observers were trained to discriminate between safe and at-risk behavior, and were then exposed to two types of interval recording procedures: momentary time sampling (MTS) and whole-interval recording (WIR). The extent and sources of observer error were measured with signal detection methods (d′ and c). The results show that observers had greater misclassification of behavior when using WIR, over MTS. The findings also revealed that observers were consistently biased toward classifying behavior as “safe” when using WIR; safe bias was also found with use of MTS, but only after previous use of WIR.     

Risk‐Based Sampling: I Don't Want to Weight in Vain

Recently, there has been considerable interest in developing risk‐based sampling for food safety and animal and plant health for efficient allocation of inspection and surveillance resources. The problem of risk‐based sampling allocation presents a challenge similar to financial portfolio analysis. Markowitz (1952) laid the foundation for modern portfolio theory based on mean‐variance optimization. However, a persistent challenge in implementing portfolio optimization is the problem of estimation error, leading to false “optimal” portfolios and unstable asset weights. In some cases, portfolio diversification based on simple heuristics (e.g., equal allocation) has better out‐of‐sample performance than complex portfolio optimization methods due to estimation uncertainty. Even for portfolios with a modest number of assets, the estimation window required for true optimization may imply an implausibly long stationary period. The implications for risk‐based sampling are illustrated by a simple simulation model of lot inspection for a small, heterogeneous group of producers.     

The impact of manufacturing and supply chain improvement initiatives: A survey comparing make-to-order and make-to-stock firms

Manufacturing firms aim at improving both internal and external processes to improve the competitive advantage. Such initiatives include lean practices as well as supplier rationalization and integration. In this paper, we analyze these improvement initiatives and their impact on business performance. In particular, we explore potential differences between make-to-order (MTO) and make-to-stock (MTS) firms. We use data from 216 Australian manufacturing firms. We find a clear difference of improvement focus between MTO and MTS firms. MTO firms exhibit a significant impact of supplier integration on business performance, but not for lean practices and supplier rationalization. The situation is completely reversed for MTS firms, since they have significant effects for internal lean practices and supplier rationalization, but not for logistics integration with supplier. The results show that the distinction between MTO and MTS firms is important when analyzing manufacturing and supply chain improvement initiatives.     

基于系统柔性的MTS-MTO混合生产决策

研究生产商采用MTS、MTO混合作业的方式为不同客户提供产品和服务的策略。计划利用一组可灵活控制的动态设备处理那些不同需求的MTS和MTO生产业务,为此,我们开发了一个多服务台的排队模型,利用拟生灭过程和相位型分布得到了MTS、MTO排队系统平衡条件和稳态概率矩阵几何解。通过求解分块矩阵方程组,给出了系统队列长度、平均等待队长、顾客服务水平等绩效测度指标。建立了系统运作成本最优化的数学模型,采用搜索算法,确定了关键参数的边界值,找到了混合系统运作的最优策略。数值模拟和系统绩效比较分析结果显示：（1）动态切换策略能够更快速的帮助MTS恢复目标库存量,控制系统缺货风险,降低库存持有成本;（2）找到了满足顾客服务水平的最少的设备配置数量和库存成本最低的生产切换时间,且动态系统的平均队列长度低于静态系统;（3）混合运作策略减少了约2/3的静态系统平均队列长度,企业在队列长度减小的窗口期内可以接受更多订单和缩短MTO订单交货时间。     

Reduce rework,improve safety: an empirical inquiry into the precursors to error in construction

A positive association between rework and safety events that arise during the construction process has been identified. In-depth semi-structured interviews with operational and project-related employees from an Australian construction organisation were undertaken to determine the precursors to rework and safety events. The analysis enabled the precursors of error to examined under the auspices of: (1) People, (2) Organisation, and (3) Project. It is revealed that the precursors to error for rework and safety incidents were similar. A conceptual framework to simultaneously reduce rework and safety incidents is proposed. It is acknowledged that there is no panacea that can be used to prevent rework from occurring, but from the findings presented indicate that a shift from a position of ‘preventing’ to ‘managing’ errors is required to enable learning to become an embedded feature of an organisation’s culture. As a consequence, this will contribute to productivity and performance improvements being realised.     

Decision Making for Risk Management: A Comparison of Graphical Methods for Presenting Quantitative Uncertainty

Previous research has shown that people err when making decisions aided by probability information. Surprisingly, there has been little exploration into the accuracy of decisions made based on many commonly used probabilistic display methods. Two experiments examined the ability of a comprehensive set of such methods to effectively communicate critical information to a decision maker and influence confidence in decision making. The second experiment investigated the performance of these methods under time pressure, a situational factor known to exacerbate judgmental errors. Ten commonly used graphical display methods were randomly assigned to participants. Across eight scenarios in which a probabilistic outcome was described, participants were asked questions regarding graph interpretation (e.g., mean) and made behavioral choices (i.e., act; do not act) based on the provided information indicated that decision‐maker accuracy differed by graphical method; error bars and boxplots led to greatest mean estimation and behavioral choice accuracy whereas complementary cumulative probability distribution functions were associated with the highest probability estimation accuracy. Under time pressure, participant performance decreased when making behavioral choices.     

Addressing the Dependency Problem in Access Security System Architecture Design

The objective of this research is to present a method for evaluating the performance of access control security systems, such as airport security operations. This requires the examination of security system architectures, which involve security technology devices and the algorithms that coordinate their operations. Dependence between device responses in multiple-device systems is a critical practical issue in assessing the performance of such architectures, though no results on this problem have appeared in the literature. This paper presents a method for evaluating when multipledevice security systems with overlapping capabilities are cost-effective. This is achieved using a dependency structure for security system devices to quantify how various technologies interact and to measure the impact of device dependence on system error probabilities. A measure of device response dependence for a two-device system is defined and its properties are explored, including bounds on the dependency measure. The effect of dependence on the system Type I and Type II error probabilities is examined for the two-device system. System performance is compared for independent vs. dependent device responses and desirable dependence relationships are identified. Results are also presented for a cascading sequence of devices. An example is presented to illustrate the results for the two-device system. Implications of these results are discussed, such as how they can be used to identify the optimal use of security devices and to determine whether new technologies warrant investment.     

Identification and Estimation of Regression Models with Misclassification

This paper studies the problem of identification and estimation in nonparametric regression models with a misclassified binary regressor where the measurement error may be correlated with the regressors. We show that the regression function is nonparametrically identified in the presence of an additional random variable that is correlated with the unobserved true underlying variable but unrelated to the measurement error. Identification for semiparametric and parametric regression functions follows straightforwardly from the basic identification result. We propose a kernel estimator based on the identification strategy, derive its large sample properties, and discuss alternative estimation procedures. We also propose a test for misclassification in the model based on an exclusion restriction that is straightforward to implement.     

Bias‐Corrected Estimation in Continuous Sampling Plans

Continuous sampling plans (CSPs) are algorithms used for monitoring and maintaining the quality of a production line. Although considerable work has been done on the development of CSPs, to our knowledge, there has been no corresponding effort in developing estimators with good statistical properties for data arising from a CSP inspection process. For example, information about the failure rate of the process will affect the management of the process, both in terms of selecting appropriate CSP parameters to keep the failure rate after inspection at a suitable level, and in terms of policy, for example, whether the process should be completely inspected, or shut down. The motivation for this exercise was developing sampling protocols for Australia's Department of Agriculture and Water Resources for monitoring the biosecurity compliance of incoming goods at international borders. In this study, we show that maximum likelihood estimation of the failure rate under a sampling scheme can be biased depending on when estimation is performed, and we provide explicit expressions for the main contribution of the bias under various CSPs. We then construct bias‐corrected estimators and confidence intervals, and evaluate their performance in a numerical study.     

金融资产收益非对称性的多标度分形测度及其在VaR计算中的应用

通过提炼多标度分形分析过程中所产生的对描述金融资产收益非对称特征有益的统计信息,提出了一种新的资产收益非对称测度--多标度分形非对称测度(Multifractal asymmetry measurement)Δf,并以沪深300指数长达7年左右的5分钟高频数据为实证样本,通过两种不同的VaR后验分析(Backtesting analysis)方法,实证对比了Δf测度和传统的偏度系数(Coefficient of skewness)测度在市场风险计算准确性方面的差异。实证结果表明:基于Δf测度的市场风险计算模型的VaR计算精度优于基于偏度系数测度的对应模型,Δf测度具有较偏度系数测度更为优异的对金融资产收益非对称特征的刻画能力。     

变动均值和方差资产配置模型的应用研究

本文在均值-方差模型的基础上,以改善估计误差为主线,选取了10种变动均值-方差的资产配置模型,以等权重策略为基准,运用了确定性等价收益和Alpha值为判断准则,同时考虑了允许卖空限制和非允许卖空的情况,实证研究结果表明：虽然在资本市场中配置模型并不能显著战胜等权重策略,但随着投资范围的扩大,模型开始显现配置效果,尤其在Alpha准则下,变动均值-方差资产配置显著。同时本文还将实证结果和目前我国投资者的实际资产配置情况进行了比较,发现了现实配置结构中的不合理之处,并提出了相应的改善建议。最后对4类常用资产进行了模拟研究,其结果也进一步证实了本文的结论。     

Alternative Models for Moment Inequalities

Behavioral choice models generate inequalities which, when combined with additional assumptions, can be used as a basis for estimation. This paper considers two sets of such assumptions and uses them in two empirical examples. The second example examines the structure of payments resulting from the upstream interactions in a vertical market. I then mimic the empirical setting for this example in a numerical analysis which computes actual equilibria, examines how their characteristics vary with the market setting, and compares them to the empirical results. The final section uses the numerical results in a Monte Carlo analysis of the robustness of the two approaches to estimation to their underlying assumptions.     

The Role of Distrust in Offshore Safety Performance

Trust is recognized as a potentially important factor in safety within high-risk industries. However, little detailed empirical research has explored how trust operates in these contexts to influence worker safety performance. The present study addresses this by (i) identifying the target (occupational group) in which trust is most important for good safety, and (ii) establishing the "type" of trust (trust or distrust) with the greatest impact on safety performance. A questionnaire survey of 203 UK offshore gas workers' attitudes of trust and distrust toward four occupational groups (workmates, supervisors, offshore managers, and contractors) and an operating company was conducted. Logistic regression analysis identified attitudes toward offshore management as the strongest predictor of safety performance at an industry level. At an installation level, safety performance was best predicted by attitudes toward contractors and workmates. Further analysis revealed attitudes of distrust as better predictors of safety performance compared to attitudes of trust. These findings suggest that safety professionals should pay more attention to the role of distrust in safety performance. They also suggest that safety initiatives should target attitudes toward specific groups for optimal effectiveness.     

考虑微观结构噪声与测量误差的波动率预测

以测量误差的分布理论为基础，本文将微观结构噪声的影响引入到测量误差的方差中，构建了包含微观结构噪声影响的HARQ-N模型。使用蒙特卡洛模拟与中国股市的高频数据对HAR、HARQ、HARQ-N模型与HAR-RV-N-CJ模型的估计和预测进行了比较，研究发现，HARQ模型和HARQ-N模型的测量误差修正项对波动率的影响系数统计显著为负，HARQ-N模型的测量误差项影响系数远大于HARQ模型，更大程度地减弱当期微观结构噪声和测量误差的影响。并且，考虑微观结构噪声和测量误差的HARQ-N模型样本内和样本外预测效果在统计上显著优于HAR模型、HARQ模型与HAR-RV-N-CJ模型。     

Causes of accidents and the time of day

Abstract

The relationship between accidents and the time of day is not entirely clear. Although there is considerable evidence that performance is poorest in the early morning hours compared to at all other times, the degree to which this is translated into accidents is not understood. In this study work-related fatal accidents were analysed to determine whether the causes of these accidents vary with time. The analysis used a classification and coding system devised to describe the wider circumstances of the causes of fatalities. In terms of absolute numbers, fatalities were most common in the late morning or early afternoon. When these results were expressed in terms of the estimated number of workers at work, the proportion of fatal accidents occurring at night was more than double that occurring during the day. Behavioural factors were the most common cause of fatalities at all times, but most common in the early hours of the morning. Further analysis of the nature of the behavioural involvement showed that errors in automatic processing (skill-based errors) were the most common types of error and did not vary much with time of day. Rule-based errors were most common during the day whereas knowledge-based errors occurred mainly in the afternoon and night shifts. These results cast some doubt on the assumption that there is a direct relationship between the previously observed variations in performance across time and the ultimate safety outcome.     

A REGRESSION ESTIMATION METHOD FOR COLLINEAR PREDICTORS

This paper addresses the important and pervasive condition of multicollinear predictors in regression analysis. The paper discusses the rudiments of a recently proposed “improved estimation” procedure (ridge regression analysis) to be used when the predictors are not orthogonal and compares the empirical performance of the ridge technique in estimating some known coefficients with the traditional estimation methods of least squares and stepwise regression.     

Effect of forecast errors on rolling horizon master production schedule cost performance for various replanning intervals

Master production schedules are usually updated by the use of a rolling schedule. Previous studies on rolling schedules seem to form the consensus that frequent replanning of a master production schedule (MPS) can increase costs and schedule instability. Building on previous research on rolling schedules, this study addresses the impact of overestimation or underestimation of demand on the rolling horizon MPS cost performance for various replanning frequencies. The MPS model developed in this paper is based on actual data collected from a paint company. Results indicate that under both the forecast errors conditions investigated in this study, a two-replanning interval provided the best MPS cost performance for this company environment. However, results from the sensitivity analysis performed on the MPS model indicate that when the setup and inventory carrying costs are high, a 1-month replanning frequency (frequent replanning) seems more appropriate for both of the above forecast error scenarios.     

套期保值, 估计风险与贝叶斯统计——基于中国铜期货市场的经验研究

针对期货最优套期保值策略估计中可能存在的估计风险问题,本文对单变量线性回归模型(OLS模型)和多变量线性回归模型(VAR模型和EC-VAR模型)进行贝叶斯分析,并采用Gibbs抽样方法对中国铜期货市场的最优套期保值策略进行了实证分析。本文还同时估计了基于频率统计方法的最优套期保值策略,并对贝叶斯统计下和频率统计下的最优套期保值策略进行了分析比较。实证结果清楚表明,估计风险对模型结果有重要影响。在处理估计风险方面,贝叶斯统计较频率统计方法有明显优势。     

一种基于改进马田系统的不平衡数据分类方法

在分类问题中,类别不平衡问题将引起分类器训练偏差,导致少数类样本诊断敏感性降低.马田系统是一种多元数据诊断和预测技术,它通过构建一个连续的测量尺度而非直接对训练样本进行学习,该性质有望不受数据分布的影响,克服分类不平衡问题.本文针对马田系统阈值计算缺陷和不平衡数据分类要求,研究一种概率阈值模型计算马田系统阈值;还针对马田系统的若干不足,采用优化模型替代正交表和信噪比筛选关键变量,并使用了一种全方位优化算法求解.通过对8个UCI数据集的实验分析表明,改进的马田系统不仅对不平衡数据有较好的分类效果,且能筛选关键变量,降维效果明显.     

SPECIFICATION AND ESTIMATION OF LOGICALLY CONSISTENT LINEAR MODELS

The implications of constrained dependent and independent variables for model parameters are examined. In the context of linear model systems, it is shown that polyhedral constraints on the dependent variables will hold over the domain of the independent variables when a set of polyhedral constraints is satisfied by the model parameters. This result may be used in parameter estimation, in which case all predicted values of the dependent variables are consistent with constraints on the actual values. Also, the implicit constraints that define the set of parameters for many commonly used linear stochastic models with an error term yield values of the dependent variables consistent with the explicit constraints. Models possessing these properties are termed “logically consistent”.