Building a Model Using Bayesian Network for Assessment of Posterior Probabilities of Falling From Height at Workplaces
Background: Falls from height are one of the main causes of fatal occupational injuries. The objective of this study was to present a model for estimating occurrence probability of falling from height. Methods: In order to make a list of factors affecting falls, we used four expert group's...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Tabriz University of Medical Sciences
2014-12-01
|
| Series: | Health Promotion Perspectives |
| Subjects: | |
| Online Access: | http://journals.tbzmed.ac.ir/HPP/Manuscript/HPP-4-187.pdf |
| Summary: | Background: Falls from height are one of the main causes of fatal occupational
injuries. The objective of this study was to present a model for estimating occurrence
probability of falling from height.
Methods: In order to make a list of factors affecting falls, we used four expert
group's judgment, literature review and an available database. Then the validity
and reliability of designed questionnaire were determined and Bayesian networks
were built. The built network, nodes and curves were quantified. For network
sensitivity analysis, four types of analysis carried out.
Results: A Bayesian network for assessment of posterior probabilities of falling
from height proposed. The presented Bayesian network model shows the interrelationships
among 37 causes affecting the falling from height and can calculate
its posterior probabilities. The most important factors affecting falling were
Non-compliance with safety instructions for work at height (0.127), Lack of
safety equipment for work at height (0.094) and Lack of safety instructions for
work at height (0.071) respectively.
Conclusion: The proposed Bayesian network used to determine how different
causes could affect the falling from height at work. The findings of this study
can be used to decide on the falling accident prevention programs. |
|---|---|
| ISSN: | 2228-6497 2228-6497 |