A hybrid model for human factor analysis in process accidents: FBN-HFACS

Human factors are the largest contributing factors to unsafe operation of the chemical process systems. Conventional methods of human factor assessment are often static, unable to deal with data and model uncertainty, and to consider independencies among failure modes. To overcome the above limitations, this paper presents a hybrid dynamic human factor model considering Human Factor Analysis and Classification System (HFACS), intuitionistic fuzzy set theory, and Bayesian network. The model is tested on accident scenarios which have occurred in a hot tapping operation of a natural gas pipeline. The results demonstrate that poor occupational safety training, failure to implement risk management principles, and ignoring reporting unsafe conditions were the factors that contributed most failures causing accident. The potential risk-based safety measures for preventing similar accidents are discussed. The application of the model confirms its robustness in estimating impact rate (degree) of human factor induced failures, consideration of the conditional dependency, and a dynamic and flexible modelling structure.     

Reflections on the use of Bayesian belief networks for adaptive management

A broad range of tools are available for integrated water resource management (IWRM). In the EU research project NeWater, a hypothesis exists that IWRM cannot be realised unless current management regimes undergo a transition toward adaptive management (AM). This includes a structured process of learning, dealing with complexity, uncertainty etc. We assume that it is no longer enough for managers and tool researchers to understand the complexity and uncertainty of the outer natural system-the environment. It is just as important, to understand what goes on in the complex and uncertain participatory processes between the water managers, different stakeholders, authorities and researchers when a specific tool and process is used for environmental management. The paper revisits a case study carried out 2001-2004 where the tool Bayesian networks (BNs) was tested for groundwater management with full stakeholder involvement. With the participation of two researchers (the authors) and two water managers previously involved in the case study, a qualitative interview was prepared and carried out in June 2006. The aim of this ex-post evaluation was to capture and explore the water managers' experience with Bayesian belief networks when used for integrated and adaptive water management and provide a narrative approach for tool enhancement.     

Particulate matter exchange between atmosphere and roads surfaces in urban areas

The deposition and the re-suspension of particulate matter (PM) in urban areas are the key processes that contribute not only to stormwater pollution, but also to air pollution. However, investigation of the deposition and the re-suspension of PM is challenging because of the difficulties in distinguishing between the resuspended and the deposited PM. This study created two Bayesian Networks (BN) models to explore the deposition and the re-suspension of PM as well as the important influential factors. The outcomes of BN modelling revealed that deposition and re-suspension of PM10 occurred under both, high-traffic and low-traffic conditions, and the re-suspension of PM2.5 occurred under low-traffic conditions. The deposition of PM10 under low-volume traffic condition is 1.6 times higher than under high-volume traffic condition, which is attributed to the decrease in PM10 caused by relatively higher turbulence under high-volume traffic conditions. PM10 is more easily resuspended from road surfaces compared to PM2.5 as the particles which larger than the thickness of the laminar airflow over the road surface are more easily removed from road surfaces. The increase in wind speed contributes to the increase in PM build-up by transporting particulates from roadside areas to the road surfaces and the airborne PM2.5 and PM10 increases with the increase in relative humidity. The study outcomes provide a step improvement in the understanding of the transfer processes of PM2.5 and PM10 between atmosphere and urban road surfaces, which in turn will contribute to the effective design of mitigation measures for urban stormwater and air pollution.     

Sex‐specific and dose‐response relationships of urinary cobalt and molybdenum levels with glucose levels and insulin resistance in U.S. adults

Growing studies have linked metal exposure to diabetes risk. However, these studies had inconsistent results. We used a multiple linear regression model to investigate the sex-specific and dose-response associations between urinary metals (cobalt (Co) and molybdenum (Mo)) and diabetes-related indicators (fasting plasma glucose (FPG), hemoglobin A1c (HbA1c), homeostasis model assessment for insulin resistance (HOMA-IR), and insulin) in a cross-sectional study based on the United States National Health and Nutrition Examination Survey. The urinary metal concentrations of 1423 eligible individuals were stratified on the basis of the quartile distribution. Our results showed that the urinary Co level in males at the fourth quartile (Q4) was strongly correlated with increased FPG (β = 0.61, 95% CI: 0.17–1.04), HbA1c (β = 0.31, 95% CI: 0.09–0.54), insulin (β = 8.18, 95% CI: 2.84–13.52), and HOMA–IR (β = 3.42, 95% CI: 1.40–5.44) when compared with first quartile (Q1). High urinary Mo levels (Q4 vs. Q1) were associated with elevated FPG (β = 0.46, 95% CI: 0.17–0.75) and HbA1c (β = 0.27, 95% CI: 0.11–0.42) in the overall population. Positive linear dose-response associations were observed between urinary Co and insulin (P_nonlinear = 0.513) and HOMA–IR (P_nonlinear = 0.736) in males, as well as a positive linear dose-response relationship between urinary Mo and FPG (P_nonlinear = 0.826) and HbA1c (P_nonlinear = 0.376) in the overall population. Significant sex-specific and dose-response relationships were observed between urinary metals (Co and Mo) and diabetes-related indicators, and the potential mechanisms should be further investigated.     

Application of Bayesian regularized BP neural network model for analysis of aquatic ecological data-A case study of chlorophyll-a prediction in Nanzui water area of Dongting Lake

Bayesian regularized BP neural network（BRBPNN） technique was applied in the chlorophyll-α prediction of Nanzui water area in Dongting Lake. Through BP network interpolation method, the input and output samples of the network were obtained. After the selection of input variables using stepwise/multiple linear regression method in SPSS i1.0 software, the BRBPNN model was established between chlorophyll-α and environmental parameters, biological parameters. The achieved optimal network structure was 3-11-1 with the correlation coefficients and the mean square errors for the training set and the test set as 0.999 and 0.000？8426, 0.981 and 0.0216 respectively. The sum of square weights between each input neuron and the hidden layer of optimal BRBPNN models of different structures indicated that the effect of individual input parameter on chlorophyll- α declined in the order of alga amount 〉 secchi disc depth（SD） 〉 electrical conductivity （EC）. Additionally, it also demonstrated that the contributions of these three factors were the maximal for the change of chlorophyll-α concentration, total phosphorus（TP） and total nitrogen（TN） were the minimal. All the results showed that BRBPNN model was capable of automated regularization parameter selection and thus it may ensure the excellent generation ability and robustness. Thus, this study laid the foundation for the application of BRBPNN model in the analysis of aquatic ecological data（chlorophyll-α prediction） and the explanation about the effective eutrophication treatment measures for Nanzui water area in Dongting Lake.     

TEMPORAL CHANGES IN THE YADKIN RIVER FLOW VERSUS SUSPENDED SEDIMENT CONCENTRATION RELATIONSHIP1

ABSTRACT: Dynamic linear models (DLM) and seasonal trend decomposition (STL) using local regression, or LOESS, were used to analyze the 50‐year time series of suspended sediment concentrations for the Yadkin River, measured at the U.S. Geological Survey station at Yadkin College, North Carolina. A DLM with constant trend, seasonality, and a log₁₀ streamflow regressor provided the best model to predict monthly mean log₁₀ suspended sediment concentrations, based on the forecast log likelihood. Using DLM, there was evidence (odds approximately 69:1) that the log₁₀ streamflow versus log₁₀ suspended sediment concentration relationship has changed, with an approximate 20 percent increase in the log₁₀ streamflow coefficient over the period 1981 to 1996. However, sediment concentrations in the Yadkin River have decreased during the decade of the 1990s, which has been accompanied by a concomitant increase in streamflow variability. Although STL has been shown to be a versatile trend analysis technique, DLM is shown to be more suitable for discovery and inference of structural changes (trends) in the model coefficient describing the relationship between flow and sediment concentration.     

Using long-term data for a reintroduced population to empirically estimate future consequences of inbreeding

Inbreeding depression is an important long-term threat to reintroduced populations. However, the strength of inbreeding depression is difficult to estimate in wild populations because pedigree data are inevitably incomplete and because good data are needed on survival and reproduction. Predicting future population consequences is especially difficult because this also requires projecting future inbreeding levels and their impacts on long-term population dynamics, which are subject to many uncertainties. We illustrate how such projections can be derived through Bayesian state-space modeling methods based on a 26-year data set for North Island Robins (Petroica longipes) reintroduced to Tiritiri Matangi Island in 1992. We used pedigree data to model increases in the average inbreeding level (F ) over time based on kinship of possible breeding pairs and to estimate empirically N_e/N (effective/census population size). We used multiple imputation to model the unknown components of inbreeding coefficients, which allowed us to estimate effects of inbreeding on survival for all 1458 birds in the data set while modeling density dependence and environmental stochasticity. This modeling indicated that inbreeding reduced juvenile survival (1.83 lethal equivalents [SE 0.81]) and may have reduced subsequent adult survival (0.44 lethal equivalents [0.81]) but had no apparent effect on numbers of fledglings produced. Average inbreeding level increased to 0.10 (SE 0.001) as the population grew from 33 (0.3) to 160 (6) individuals over the 25 years, giving a ratio of 0.56 (0.01). Based on a model that also incorporated habitat regeneration, the population was projected to reach a maximum of 331–1144 birds (median 726) in 2130, then to begin a slow decline. Without inbreeding, the population would be expected stabilize at 887–1465 birds (median 1131). Such analysis, therefore, makes it possible to empirically derive the information needed for rational decisions about inbreeding management while accounting for multiple sources of uncertainty.     

Tracking a half century of media reporting on gray wolves

Natural resource and wildlife managers must balance the disparate priorities of a diversity of stakeholders. To manage these priorities, a firm understanding of topics salient to the public is needed. The media often report on issues of importance to the public; therefore, these reports may be a useful measure of public interest. However, efficient methods for distinguishing diverse topics related to a wildlife management issue reported in the media and changes in the salience of those topics have been lacking. We used latent Dirichlet allocation, a Bayesian mixture model, to quantitatively assess the salience of topics surrounding the gray wolf (Canis lupus), which was reintroduced to Idaho (U.S.A.) in 1995. We analyzed articles published from 1960 to 2015 in an Idaho newspaper. We identified 6 distinct topics associated with gray wolves: policy, hunting, biological status, implementation of management, recovery, and human-wolf conflict. The salience of topics pre- and postreintroduction of wolves (1995) and pre- and postdelisting of wolves from the U.S. Endangered Species Act (2009) differed significantly, underscoring that these events were turning points in how issues were being publicly discussed and framed. Articles written by the local reporters were more likely to report on topics regarding conflict between humans and wolves, whereas articles sourced from a national outlet reported more on topics pertaining to wolf policy and biological status. In the context of managing a contentious, far-ranging, and long-lived wildlife species, our methods can help guide the location and timing of a suite of management strategies (e.g., media relation plans and stakeholder engagement) that promote human-wildlife coexistence across different landscapes.     

A Bayesian network-based approach for the assessment and management of ageing in major hazard establishments

Currently, there is an increasing attention towards ageing of industrial equipment, as the phenomenon has been recognised as a cause of severe accidents, recorded in the last years in many process establishments. Recent studies described ageing through a number of key-factors affecting the phenomenon by accelerating or slowing it down. The Italian Competent Authority for the prevention of chemical accidents (Seveso III Directive) adopted a short-cut method, accounting for the assessment of these factors, to evaluate the adequateness of ageing management during inspections at Seveso sites. In this paper, a Bayesian Network was developed, by using the data gathered during the first application of the short-cut method, with the aim to verify the robustness of the approach for ageing assessment and the validity of the a priori assumptions used in assessing the key-factors. The structure of the Bayesian network was established by using experts’ knowledge, whereas the Counting Learning algorithm was adopted to execute the parameter learning by means of the software Netica. The results showed that this network could effectively explore the complex logical and uncertain relationships amongst factors affecting equipment ageing. Results of the present study were exploited to improve the short-cut method.     

Fault detection and diagnostics of a three-phase separator

A high demand of oil products on daily basis requires oil processing plants to work with maximum efficiency. Oil, water and gas separation in a three-phase separator is one of the first operations that are performed after crude oil is extracted from an oil well. Failure of the components of the separator introduces the potential hazard of flammable materials being released into the environment. This can escalate to a fire or explosion. Such failures can also cause downtime for the oil processing plant since the separation process is essential to oil production. Fault detection and diagnostics techniques used in the oil and gas industry are typically threshold based alarm techniques. Observing the sensor readings solely allows only a late detection of faults on the separator which is a big deficiency of such a technique, since it causes the oil and gas processing plants to shut down.A fault detection and diagnostics methodology for three-phase separators based on Bayesian Belief Networks (BBN) is presented in this paper. The BBN models the propagation of oil, water and gas through the different sections of the separator and the interactions between component failure modes and process variables, such as level or flow monitored by sensors installed on the separator. The paper will report on the results of the study, when the BBNs are used to detect single and multiple failures, using sensor readings from a simulation model. The results indicated that the fault detection and diagnostics model was able to detect inconsistencies in sensor readings and link them to corresponding failure modes when single or multiple failures were present in the separator.