Bayesian population

Jonsson F, Johanson G. 2003. The Bayesian population approach to physiological toxicoki-netic-toxicodynamic models—an example using the MCSim software. Toxicol Lett 138 143-150. 

Yang RSH, Mayeno AN, Lyons M, Reisfeld B. 2010. The application of physiologically-based pharmacokinetics (PBPK), Bayesian population PBPK modeling, and biochemical reaction network (BRN) modeling to chemical mixture toxicology. In Mumtaz M, editor, Principles and practices of mixture toxicology. Hoboken (NJ) John Wiley Sons. 

The simplest method of accounting for variability consists in identifying the parameters, experimental data, and model output with an average or reference individual [45], One shortcoming of this approach is that although general model behavior is representative, much of the actual population may not be well described. Methods attempting to explicitly account for variability and uncertainty include Monte Carlo simulations [18,46,48], Bayesian population methods [49-51,53,56], the use of fuzzy sets [44,52], and other probability based methods [43], Monte Carlo simulations, which model the parameter variability in terms of probability distributions, are the most common methods. Each individual is characterized by a set of parameters whose values are drawn from a... 

Jonsson F. Development of Bayesian population models. In Marklund S, editor. Physiologically Based Pharmacokinetic Modeling in Risk Assessment. 

Jonsson F. Physiologically based pharmacokinetic modeling in risk assessment. In Development of Bayesian population methods. Stockholm Uppsala University, 2001. 

Jonsson, E., and Johanson, G. (2002). Physiologically based modeling of the inhalation kinetics of styrene in humans using a bayesian population approach. Toxicol Appl Pharmacol 179, 35 9. 

A Bayesian population variability analysis for estimation of the work time loss distributions due to occupational accidents... 

ABSTRACT Occupational accidents pose several negative consequences to employees, employers, environment and people surrounding the local where the accident takes place. Thus, this paper proposes a Bayesian population variability analysis-based method for the prediction of work time loss distributions due to occupational accidents. The use of Bayesian analysis aims at investigating future trends regarding occupational accidents in the workplace as well as enabling a better management of the labor force. In fact, the key performance indicators here estimated are the expected unavailability of the labor force and also the expected recovery time from an accident which will be computed by a Markov-based model. Finally, the paper presents a practical example by using available data on occupational accidents from a hydroelectric power company in Brazil to predict the work time losses and their respective distributions. 

Thus, this paper proposes a Bayesian population variability analysis-based method for this... 

The remainder of this paper is organized as follows. Section 2 presents the theoretical background about the Bayesian Population Variability Analysis. Section 3 presents the proposed model, while Section 4 discusses the numerical results using evidence from a real accident database of a hydroelectric power company in Brazil. Finally, section 5 presents concluding remarks. 

Bayesian population variability assessment for accident analysis... 

The Bayesian population variability analysis of p assumes that the distribution, 0,) is... 

In this section, we illustrate and discuss the use of the Bayesian Population Variability analysis and Markov-based models by means of an example. We start by supposing that we are interested in assessing the average distribution of work time loss due to occupational accidents for workers of a hydroelectric power company in Brazil. Runtime data were collected from the timeline of operation employees between 01/01/2005 and 09/31/2012 in order to construct the likelihood function. 

As discussed in Section 2.1 the Bayesian Population Variability Analysis must be used in a workers population subject to similar risk of accidents. So, in this section, the model is applied to employees with the same job and workplace. The administrators located in the operation sector of the company were analyzed. These workers suffered mainly two types of accidents (1) accidents in commuting and (2) falls. It is expected that one worker has different accident and recovery rates for each type of accident. Therefore, the model was applied separately for analyzing the work time loss distributions due to accidents in commuting and falls. 

To validate the application of the model we generate the data from the results of the Bayesian population variability model in order to compare them with the actual data. The Bayesian model applied in the real case provides estimations for population variability distributions of accidents and recoveries rates. From these distributions we generate the accidents and recoveries data of workers. It is expected that these data can represent the accident-recovery process of this population of workers. 

The model presented in this paper represents an extension of the Bayesian population variability assessment method in accidents analysis. A Markov-based model is used for estimation of the expected work time loss distributions due to occupational accidents. The Bayesian population variability assessment method allows the evaluating of population variability of accidents and recovery rates based on exposure data of workers submitted to same occupational risks and the Markov-based model is used to derive the worker unavailability... 

KeRy, M. Schaub, M. 2012. Bayesian population analysis using WinBUGS a hierarchical perspective, Boston, Academic Press. 

---