Resampling methods of analysis in simulation studies

Cheng, Wood-Hi; Currie, Christine S. M.

doi:10.1109/wsc.2009.5429319

Cited by 8 publications

(7 citation statements)

References 4 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Due to the complexity of the system under consideration, nonlinear dynamics methods are used to detect fires in the form of a modified method for calculating recurrence plots for the current values of the vector of increments of the state of the gaseous medium. To assess the reliability of detecting fires in the premises based on the use of the current measure of the recurrence of the vector of increments of hazardous parameters of the gas environment, the method of the sample distribution function was used [37,38].…”

Section: The Study Materials and Methodsmentioning

confidence: 99%

Development of a method for assessing the reliability of fire detection in premises

Sadkovyi

Поспелов²,

Rybka

et al. 2022

EEJET

View full text Add to dashboard Cite

The object of this study is the detection of fires in the premises. The problem that was solved is the development of tools to assess the reliability of detection of fires in the premises based on the recurrence of the vector of increases in dangerous parameters of the gas environment. The method includes the sequential implementation of five procedures related to the formation of the vector of current increases in dangerous parameters, determining the recurrence of the current vector and evaluating the empirical distribution function relative to the calculated current recurrence of the state vector. Features and distinctive attributes of the developed method are the use of empirical cumulative distribution function in relation to the current recurrence of the state of hazardous parameters of the gas environment in the premises during fires. This makes it possible to solve the task of developing tools for the numerical determining of the trust limit for the predefined level of significance (reliability) and the likelihood of detecting fires in the premises in real time. The scope and conditions for the practical use of the obtained results are the modern and promising means and fire protection systems of various types of premises in buildings and structures. The proposed method was tested on the example of igniting test materials in the laboratory chamber. It is established that for materials with a high combustion rate (alcohol and cellulose) with a probability of 0.95, there is a sharp decrease in the value of the empirical function assessment to zero values. For timber, the value of this estimate is 0.15, and for textiles, the minimum value of the estimate is 0.31. It is established that the boundaries of the confidence interval with the level of significance covering the obtained estimates are determined by the value of ±0.086. In general, the results of the test indicate the operability of the proposed method for determining the reliability of detection of fires in the premises on the basis of the current degree of recurrence of increases in dangerous parameters of the gas environment

show abstract

Section: The Study Materials and Methodsmentioning

confidence: 99%

Development of a method for assessing the reliability of fire detection in premises

Sadkovyi

Поспелов²,

Rybka

et al. 2022

EEJET

View full text Add to dashboard Cite

show abstract

“…In practice, it is really a big problem that large numbers of simulation replications may be too impractical to do. Bootstrap method is a very general method for numerically estimating the distribution of a statistics [9]. It is based on the following idea of replacing the distribution of the sample by the estimate we have for it.…”

Section: The Bootstrap Methodsmentioning

confidence: 99%

Statistical Analysis and Verification of the Satellite Line of Sight Pointing Jitter

Dong

Wang

Zhang

et al. 2010

2010 First International Conference on Pervasive Computing, Signal Processing and Applications

View full text Add to dashboard Cite

A method of statistical analysis and verification of the satellite line of sight jitter is discussed. For the case that the jitter measurement is random, the verification of jitter is achieved by estimate the probability that the jitter requirement is not met. The binary analysis method of simulation result is introduced and the miss rate which can be calculated from the result of the binary analysis is proposed for the fact that the statistical mean of the miss rate is an estimate of the probability of jitter requirement misses. The distribution of the miss rate is discussed for the purpose of estimate the statistics of miss rate. To avoid trivial simulation replications in analysis and verification, the bootstrap method is used to obtain a normally distributed miss rate sample, from which the mean miss rate and its variance can be estimated. As thus the confidence interval of miss rate is evaluated to provide a much safer verification result. The credibility and time cost of the method is discussed in case study, and an introductive example is presented to give an introduction of application of the method.

show abstract

“…We use bootstrapping for our analysis, a relatively straightforward way of dealing with non-standard data. Cheng and Currie (2009) give an introduction to the use of bootstrapping methods in the analysis of simulation models but for a more comprehensive description of analyzing non-standard data, we recommend Cheng (2017).…”

Section: R2mentioning

confidence: 99%

Practical Considerations in Selecting the Best Set of Simulated Systems

Monks¹,

Currie²

2018

2018 Winter Simulation Conference (WSC)

View full text Add to dashboard Cite

In many practical problems, simulation models are used to support complex decision-making processes comparing hundreds or thousands of solutions. These problems typically have a key objective but the final decision may be dependent on other factors, which cannot be incorporated into the simulation model. In such cases, decision-makers may request a short list of 'good' solutions, which work well for the main objective and satisfy one or more chance constraints. While fully sequential ranking and selection procedures can be effective at solving these problems, surveys of experimentation practice suggest that they are under-utilized, potentially due to difficulties automating commercial software. We develop an approach with just two stages of replications. The approach, which has been designed to cope with the use of common random numbers, draws on ideas from indifference zones and makes use of bootstrapping to find a subset of high quality solutions. A Python implementation is freely available.

show abstract

Resampling methods of analysis in simulation studies

Cited by 8 publications

References 4 publications

Development of a method for assessing the reliability of fire detection in premises

Development of a method for assessing the reliability of fire detection in premises

Statistical Analysis and Verification of the Satellite Line of Sight Pointing Jitter

Practical Considerations in Selecting the Best Set of Simulated Systems

Contact Info

Product

Resources

About