One of the authors of the study, Teresa Rivas, from the Mining Engineering School at this university explained to SINC that job duration is also related to adopting bad positions and lack of knowledge of regulations, as a result of which shorter tasks are associated with more falls and less well-trained workers. On the other hand, workers that carry out longer tasks are better prepared for dangerous situations.
To carry out this study, authors surveyed almost 400 workers from 103 companies from industry, construction and the services sector while performing their tasks at height. Bayesian networks, one of the so called “automatic learning” statistical techniques enabling graphic expression of relations of dependence between variables, as well as making estimates in probalistic terms of the influence of each variable and its interaction in the type of fall, were used to analyse researchers’ data.
In previous work these scientists have already analysed the capacity of predicting and interpreting different automatic learning techniques and concluded that Bayesian networks, apart from having a comparable predictive capacity to other statistical methods, “have an excellent interpretative capacity”. Rivas points out that Bayesian networks can be improved with new data, as a result of which they are a very useful tool for identifying causes of accidents and establishing the bases of understanding for a “solidly based” prevention policy.
José María Matías, mathematician from the Department of Statistics at the University of Vigo who also participated in the study, says that these types of techniques can be used for analysing causes “because there is continuous data on victims of work accidents, but beyond just the numbers what we need to look for is how to avoid such accidents, and this is where these methodological contributions play an important role”.
The researcher Rivas concluded that knowing the dependence relationship between the causes of the accident offers experts useful information for improving health and safety at work management models.Referencias bibliográficas:
J. M. Matías, T. Rivas, J. E. Martín, J. Taboada. A machine learning methodology for the analysis of workplace accidents. International Journal of Computer Mathematics (85):3-4, 559 – 578 Marzo-Abril 2008
SINC Team | alfa
Rutgers-led innovation could spur faster, cheaper, nano-based manufacturing
14.02.2018 | Rutgers University
New study from the University of Halle: How climate change alters plant growth
12.01.2018 | Martin-Luther-Universität Halle-Wittenberg
For the first time, a team of researchers at the Max-Planck Institute (MPI) for Polymer Research in Mainz, Germany, has succeeded in making an integrated circuit (IC) from just a monolayer of a semiconducting polymer via a bottom-up, self-assembly approach.
In the self-assembly process, the semiconducting polymer arranges itself into an ordered monolayer in a transistor. The transistors are binary switches used...
Breakthrough provides a new concept of the design of molecular motors, sensors and electricity generators at nanoscale
Researchers from the Institute of Organic Chemistry and Biochemistry of the CAS (IOCB Prague), Institute of Physics of the CAS (IP CAS) and Palacký University...
For photographers and scientists, lenses are lifesavers. They reflect and refract light, making possible the imaging systems that drive discovery through the microscope and preserve history through cameras.
But today's glass-based lenses are bulky and resist miniaturization. Next-generation technologies, such as ultrathin cameras or tiny microscopes, require...
Scientists from the University of Zurich have succeeded for the first time in tracking individual stem cells and their neuronal progeny over months within the intact adult brain. This study sheds light on how new neurons are produced throughout life.
The generation of new nerve cells was once thought to taper off at the end of embryonic development. However, recent research has shown that the adult brain...
Theoretical physicists propose to use negative interference to control heat flow in quantum devices. Study published in Physical Review Letters
Quantum computer parts are sensitive and need to be cooled to very low temperatures. Their tiny size makes them particularly susceptible to a temperature...
15.02.2018 | Event News
13.02.2018 | Event News
12.02.2018 | Event News
21.02.2018 | Materials Sciences
21.02.2018 | Health and Medicine
21.02.2018 | Physics and Astronomy