An analysis of biorhythms and their influence on motor vehicle fatalities

The possible effect of proposed cyclic fluctuations of human sensitivity (28 days), intelligence (33 days), and physical (23 days) characteristics on motor vehicle accidents and fatalities was examined. Three methods of calculating these theoretical cycles were also examined. One method, the “Biomate” computer, was found to be unreliable in that it indicated 10.6% more days designated by the theory as critical than were actually present in a 300 case sample. In a sample of 141 motor vehicle fatalities (64 drug-free and 77 drug-associated) no correlation with the theoretical critical days was found for either group beyond that expected from random distribution. No evidence was found to support the increased likelihood of accident involvement on the “critical day” as implied by the concept of the “biorhythm” theory. Several possible sources of error in previously published reports are discussed.     

Attitudes of commercial motor vehicle drivers towards safety belts

Despite the fact that Hawaii has one of the highest seat belt use rates for passenger vehicles in the United States, and has had a mandatory seat belt use law since the 1980s, studies have shown that commercial motor vehicles (CMV) seat belt use rates are low. To better understand this phenomenon, a comprehensive survey of commercial vehicle drivers was conducted in Hawaii to ascertain attitudes and self-reported behaviors regarding seat belt use. A total of 791 drivers responded to a written questionnaire implemented at weigh stations and distributed to various trucking firms and transport centers. Approximately 67% reported that they use seat belts “always” when driving a CMV (commercial motor vehicle), yet when asked how often do other CMV drivers use seat belts, only 31% responded “always.” Interestingly, 86% of these same drivers reported that they use seat belts “always” when driving a personal vehicle. The major reason cited for non-use of belts was “frequent stops/inconvenience” (29%), and “not safety conscious” (23%). Notably, the self-reported use of safety belts is highest among operators of vans (88% said “always”), followed by buses (87% said “always”) and lowest among truck drivers (only 60% said “always”). In this paper, some of the differences between self-reported users and non-users are explored and a multivariate logit model was developed to predict the odds of belt use as a function of various factors.     

A study of driver experience and vehicle familiarity in accidents

A separate study was performed as part of a major accident investigation project to (1) test the hypothesis that drivers of vehicles involved in investigated accidents had less driving experience than the general driving population, and (2) test the hypothesis that accident involved drivers were less familiar with their vehicles than the general driving population. A Kolmogorov-Smirnov one-tail test was employed to evaluate the differences in the distributions of driver experience and vehicle familiarity between the “accident sample” and a “control sample.” Results indicated that drivers involved in the investigated accidents had less driving experience in general and had less driving experience with the accident vehicles than drivers in the general population. These effects are discussed briefly and further studies are suggested.     

Multiple-vehicle traffic accidents in Hong Kong

‘Multiple-vehicle traffic accident’ refers to a crash between two or more moving objects. Unlike single-vehicle accidents, not all drivers involving in a multiple-vehicle accident are responsible for the occurrence of the event. Accordingly, variables such as road type, speed limit and number of vehicles involved in the accident are expected to play a much more important role in association with injury severity in multiple-vehicle accidents. To study the factors influencing injury severity of multiple-vehicle traffic accidents, a population-based study was conducted. The traffic accident data was obtained from the Traffic Accident Data System (TRADS), which was developed by the Transport Department, Police Force and Information Technology Services Department, Hong Kong. Multiple-vehicle traffic accidents (N = 10,630) occurring during the 2-year period 1999/2000 were considered. Potential risk factors such as district, human, vehicle, safety, environmental and site factors were examined. Categorizing injury severity into “fatal/serious” and “slight”, a stepwise logistic regression model was applied to the population data set. The district board, time of the accident, driver's gender, vehicle type, road type, speed limit and the number of vehicles involved are significant factors influencing the injury severity. Identification of risk factors for severe traffic accidents provides valuable information to help with new and improved road safety control measures.     

Runout analysis in fatigue investigation

Those individual tests in a series of experiments that are not completed or must be suspended are known as “censored points,” or “runouts”. Inclusion of runouts in data analysis can be problematic, and such ad hoc approaches as ignoring the runout observation or treating it as a failure can significantly affect estimation. The methodology offered here alleviates the handling of runouts and censored data by using the maximum likelihood estimation (MLE) method to incorporate the censored data properly. The methodology is illustrated with an example problem using actual data and the affects of ad hoc approaches are illustrated.     

The use of energy windowing to discriminate SNM from NORM in radiation portal monitors

Energy windowing is an algorithmic alarm method that can be applied to plastic scintillator-based radiation portal monitor (RPM) systems to improve operational sensitivity to certain threat sources while reducing the alarm rates from naturally occurring radioactive material. Various implementations of energy windowing have been tested and documented by industry and at Pacific Northwest National Laboratory, and are available in commercial RPMs built by several manufacturers. Moreover, energy windowing is being used in many deployed RPMs to reduce nuisance alarms and improve operational sensitivity during the screening of cargo. This paper describes energy windowing algorithms and demonstrates how these algorithms succeed when applied to “controlled” experimental measurements and “real world” vehicle traffic data.     

A typology for drinking driving offenders: Methods for classification and policy implications

Comprehensive arrest histories were obtained for 353 DUI offenders who were referred to a probation and rehabilitation demonstration program. The average number of total offenses was 7.9, with 89% of all tracked offenders having more than one offense. Approximately 63% of all recorded offenses were other types of offenses besides DUI. The overall arrest history profile of the group suggested that many DUI offenders are habitual violators of other laws as well. To identify distinctive arrest profiles within the referral sample, a Q mode factor analysis followed by a discriminant function analysis was used to classify offenders into profile subgroups. Five distinctive subgroups emerged. A “low offense” group was characterized by the lowest average number of overall arrests and also contained all offenders with no arrest besides the index DUI arrest. A “mixed” group had a higher average number of total arrests than the “low offense” group and diverse types of offenses. A young “traffic” group was distinguished by many hazardous moving violations other than DUI. Two smaller and older groups—a “public drunkenness” group and a “license” group—had the highest average number of arrests including DUI, public drunkenness, license violations, equipment violations, disturbance arrests and assault arrests. These subgroups were found to differ on demographic variables and drinker status variables. The “public drunkenness” group was found to have the highest accident rate. Groups were compared to groups found in other cluster analyses. Also, treatment implications were discussed. It was suggested that treatment programs focusing exclusively on changing alcohol consumption behavior are not likely to reduce accident risk for some of the offender groups. For example, it was suggested that effective intervention for the “traffic” group should target driving behavior, whether drunk or sober, rather than focus exclusively on consumption behavior. Other alternatives are also discussed.     

Commentary on statistical issues underlying the evaluation of a social policy change on motor vehicle accident involvement

This commentary addresses the statistical issues which are involved in assessing whether a causal interpretation can be given to an association between a change in social policy such as lowering the legal age for drinking alcohol and a change in the nature of accident involvement frequency for a particular sub-population of drivers like 18–20 year olds. In this regard, two methodological concerns are emphasized. The first is the resolution of a possible paradox in the nature of an association in the sense that individuals with a greater tendency to accident involvement may also have a greater tendency to alcohol usage so that their alcohol related accident involvement is coincidental rather than causal. For this setting, alcohol would be an “after the fact” correlate with accident involvement as opposed to a possible cause. Similarly, the second concern is the role of population exposure as an explanation for the change in alcohol related accident frequency for a particular age group. Since availability naturally increases the size of the population at risk for such accidents, the corresponding accident frequency may similarly increase even though the corresponding accident rates may have remained unchanged or perhaps even decreased. For these reasons, although associations with change in social policy are definitely of general interest, their interpretation in a causal sense should be viewed with substantial caution until the implications of competing explanations are reconciled. This assertion, however, does not mean that such associations are not useful for policy purposes, but rather that they lack sufficiently targeted statistical validity to be the primary basis for such change. In other words, if there is a strong rationale on economic, political, or social grounds for a policy change, then data presenting meaningful associations which are compatible with it can often play an important supportive role for it.     

Quantifying the influence of safety management on the reliability of safety barriers

A methodology is described that enables to use safety management audit assessments and safety culture questionnaire results for estimating the reductions in the reliability of safety barriers in major hazard plants. The critical issue is the establishment of weight factors in combination with the anchoring of “good” safety management. A method is proposed to derive weight factors from statistical accident analysis in combination with a statistical analysis of safety management assessments at a representative sample of major hazard industries. A preliminary set of weight factors is presented with some examples of resulting reductions in reliability—this demonstration confirms that the set of weight factors needs further development.     

How accurately does the public perceive differences in transport risks?: An exploratory analysis of scales representing perceived risk

This paper probes the extent to which the public accurately perceives differences in transport risks. The paper is based on a survey of a random sample of the Norwegian population, conducted in September 2003. In the survey, respondents were asked: “How safe do you think it is to travel by means of (bus, train, etc.)?” Answers were given as: very safe, safe, a little unsafe, and very unsafe. A cursory examination of the answers suggested that the Norwegian public was quite well informed about differences in the risk of accident between different modes of transport, as well as between groups formed according to age and gender for each mode of transport. This paper probes the relationship between statistical estimates of risk and summary representations of perceived risk more systematically. It is found that the differences in fatality rate between different modes of transport are quite well perceived by the Norwegian public, irrespective of the way in which perceived risk is represented numerically. The relationship between statistical estimates of risk and numerical representations of perceived risk for each mode of transport is more sensitive to the choice of a numerical representation of perceived risk. A scale in which the answer “very safe” is assigned the value of 0.01 and the answer “very unsafe” is assigned the value of 10 is found to perform quite well. When the perception of risk is represented numerically according to this scale, a positive correlation between statistically estimated risk and perceived risk is found in seven of the eight comparisons that were made to determine how well variation in accident rates according to age and gender for car occupants, car drivers, cyclists and pedestrians are perceived.     

Using traffic conviction correlates to identify high accident-risk drivers

One of the primary missions of the California Department of Motor Vehicles is to protect the public from drivers who represent unacceptably high accident risks. Optimum fulfillment of this objective requires the development and implementation of strategies for identifying high-risk drivers. One such system in California is the department’s negligent operator point system. This system assigns points to moving violations and accidents and authorizes the department to take driver control actions against drivers who meet the prima facie definition of a negligent operator. The present study explored the viability of predicting accidents from equations constructed to predict convictions for the general driving population. Equations or models that better identify drivers at increased risk of future accident involvement would increase the number of accidents prevented through post license control actions. Although the results did not support prior findings that equations keyed to citations do as well as or better than equations keyed to accidents in predicting subsequent accident involvement, a canonical correlation approach considering subsequent accident and citation rates simultaneously produced a 14.9% improvement in the classification accuracy or “hit rate” for identifying accident-involved drivers.     

Fatigue life enhancement of aluminium joints through mechanical and thermal prestressing

This paper presents some simple and flexible methods to enhance the fatigue life of welded aluminium components. Besides enhancing the fatigue life, the proposed methods can easily be implemented into manufacturing processes. The key element of the methods is to change residual stresses from tension to compression at locations vulnerable to fatigue. This is accomplished by mechanical prestressing using elastic pre-deformation or by thermal prestressing using induction heating. The specimens tested are welded aluminium rectangular hollow section T-joints. Prior to fatigue testing, welding FE-simulations were carried out to verify the magnitude and pattern of the residual stress fields (through process modeling). Fatigue testing was later carried out on four different batches. One batch was produced using elastically pre-deformed chords, two batches were treated by means of thermal prestressing (induction heating), and one batch was “as welded” representing a “reference case”. Based on statistical evaluation of S–N data, the introduction of superimposed compressive stress fields results in a significantly improved fatigue life. Among the different batches, induction heating turned out to be the most promising method with a fatigue strength improvement factor of 1.5 on stress, compared to “as welded” components.     

An Investigation of the Measurements Systems Analysis “Analytic Method” for Attribute Gages

Three automotive corporations have developed and sanctioned the recently revised reference manual entitled Measurement Systems Analysis. This “standard” contains a procedure, called the “analytic method,” whose purpose is to estimate the gage bias and gage repeatability of an attribute gage. An improved estimation procedure for this standard is presented. The improved estimation procedure yields more accurate estimates than those obtained using the procedures currently presented in the standard. In addition, the improved procedure allows more flexibility in data collection than the current test protocol. A simulation study that evaluates the estimation procedure of the current standard and compares it with the improved estimation procedure is presented. Errors that are contained in the present standard are also noted.     

Modeling vehicle accidents and highway geometric design relationships

The statistical properties of four regression models—two conventional linear regression models and two Poisson regression models—are investigated in terms of their ability to model vehicle accidents and highway geometric design relationships. Potential limitations of these models pertaining to their underlying distributional assumptions, estimation procedures, functional form of accident rate, and sensitivity to short road sections, are identified. Important issues, such as the treatment of vehicle exposure and traffic conditions, and data uncertainties due to sampling and nonsampling errors, are also discussed. Roadway and truck accident data from the Highway Safety Information System (HSIS), a highway safety data base administered by the Federal Highway Administration (FHWA), have been employed to illustrate the use and the limitations of these models. It is demonstrated that the conventional linear regression models lack the distributional property to describe adequately random, discrete, nonnegative, and typically sporadic vehicle accident events on the road. As a result, these models are not appropriate to make probabilistic statements about vehicle accidents, and the test statistics derived from these models are questionable. The Poisson regression models, on the other hand, possess most of the desirable statistical properties in developing the relationships. However, if the vehicle accident data are found to be significantly overdispersed relative to its mean, then using the Poisson regression models may overstate or understate the likelihood of vehicle accidents on the road. More general probability distributions may have to be considered.     

Estimating the number of accidents at intersections from a knowledge of the traffic flows on the approaches

The increase in the number of accidents at intersections, relative to non-junction accidents, may be explained by the basic fact that an increase in the number of vehicles on the road is generally accompanied by an increase in the number of collisions, which rise at a faster rate than single-vehicle accidents. Generally, more than 50% of the collisions occur at intersections. Data from a number of countries studied support the above statement and show that over the years, the number of intersection accidents has increased at a faster rate than other accidents. The study includes a statistical analysis of the general trends in the number of intersection accidents, their severity, and an analysis of types of accidents at intersections. A model was developed which enables the estimation of the expected number of accidents at individual intersections. It was found that vehicle exposure can be used as the basis for estimation. The exposure, denned as the number of occasions for accidents, was calculated through the sum of the products of flow at the 24 points where vehicle paths cross or merge. This measure of exposure, expressed as a traffic flow index, showed good correlation with the number of accidents. On the assumption that the number of accidents at an intersection, in a given time interval, is Poisson distributed, significance tests were made comparing the actual number of accidents with the expected according to the vehicle exposure index. Such comparisons are useful in the determination of accident “black spots”.     

On the prevention of accidents and injuries : A comparative analysis of conceptual frameworks

The meaning of prevention has changed as new applications of the concept have appeared. Ideas presented in eleven different conceptual frameworks are compared. Identification of the frameworks took place through searches in databases and relevant literature. Five are general by nature, while six relate to injuries and accidents. All are supported by just a few parameters, the time dimension being the most prominent. Compatibility was established on three additional dimensions: level (individual, organizational or societal); direction (“bottom-up” or “top-down”); and in relation to the trichotomy “host-agent-environment”. An attempt to synthesize all these dimensions into one general model of accident and injury prevention is presented.     

Suggestions for an improved system for statistical data about road accidents and road traffic

Data information systems for road accidents and road traffic must satisfy high standards of relevance and quality. The general outline of an improved system for collecting road accident data is given. The system is characterized by the use of statistical sampling methods. The police, the insurance companies and the hospitals are recommended as sources of information about the total accident population. A statistical sample of all identified accidents is then investigated in more detail by special local investigation groups. A hypothetical numerical example is given to show how the suggested system would work in practise. Road accident data should not be isolated from road traffic data. An improved system for collecting information on road traffic is also discussed. This consists of a basic system (founded on statistical sampling methods) for estimation of the total volume of traffic and a few other essential variables such as the volume divided into speed and vehicle types.     

On the use of annual vehicle miles of travel estimates from vehicle owners

Estimates of annual vehicle miles of travel from 433 vehicle owners are compared with “actual” annual mileages as obtained from odometer readings recorded as part of North Carolina's Motor Vehicle Inspection Program. The results show that owners tend to overestimate annual mileage on low-usage vehicles and underestimate annual mileage on high-usage vehicles. On the average, the owners' estimate did not differ significantly from the actual mileage. A single estimate of annual vehicle miles of travel is not reliable, as errors of estimation are large. However, the distribution of errors is approximately normal and centered near zero. Hence, the mean annual miles may be estimated with any desired precision by including more owners in the sample.     

Statistical calibration of computer simulations

This paper surveys issues associated with the statistical calibration of physics-based computer simulators. Even in solidly physics-based models there are usually a number of parameters that are suitable targets for calibration. Statistical calibration means refining the prior distributions of such uncertain parameters based on matching some simulation outputs with data, as opposed to the practice of “tuning” or point estimation that is commonly called calibration in non-statistical contexts. Older methods for statistical calibration are reviewed before turning to recent work in which the calibration problem is embedded in a Gaussian process model. In procedures of this type, parameter estimation is carried out simultaneously with the estimation of the relationship between the calibrated simulator and truth.     

Qualitative Ordinal Scales: The Concept of Ordinal Range

Many practical problems of quality control involve the use of ordinal scales. Questionnaires planned to collect judgments on qualitative or linguistic scales, whose levels are terms such as “good,” “bad,” “medium,” etc., are extensively used both in evaluating service quality and in visual controls for manufacturing industry. In an ordinal environment, the concept of distance between two generic levels of the same scale is not defined. Therefore, a population (universe) of judgments cannot be described using “traditional” statistical distributions since they are based on the notion of distance. The concept of “distribution shape” cannot be defined as well. In this article, we introduce a new statistical entity, the so-called ordinal distribution, to describe a population of judgments expressed on an ordinal scale. We also discuss which of the traditional location and dispersion measures can be used in this context and we briefly analyze some of their properties. A new dispersion measure, the ordinal range, as an extension of the cardinal range to ordinal scales, is then proposed. A practical application in the field of quality is developed throughout the article.