Effectiveness Research Of Safety Signs In Coal Mines Based On Eye Movement Experiment

Shui-cheng Tian,Lu Hui,Hong-xia Li

DOI: https://doi.org/10.1007/978-3-319-58521-5_23

2017-01-01

Abstract:Through the analysis of major fatal accidents occurred within 30 years from 1986 to 2016 in China's coal mines, it reflects that human errors accounting for up to 80% of the total accidents. The coal mines have set up a variety of safety signs that can alert miners to be careful with the insecurity in the workplace, and guide them to behave safely in accordance with legal standards. However, the human-induced accidents still occur. The problem is that although the safety signs have been adopted universally in the coal mines as a kind of management strategy, they are hard to play the safety signs' due role in the practical application. The low effectiveness of safety signs is one of the important reasons for the coal mines accidents. In order to settle the practical problems and improve the safety level of coal mines, this research is designed to study the effectiveness of safety signs in coal mines.As a result of the previous researches, the author divided safety signs in coal mines' practical application into four stages by utilize 'communication-human' information processing model and determined the effective factors of safety signs in accordance with the transfer characteristics of each stage. Thus the author had already established the effective factors system of safety signs and determined the weight of factors that influence the safety signs' effectiveness by using the method of Analytic Hierarchy Process. The paper selected the setting positions of safety signs according to the importance ranking of safety signs' effectiveness factors.In order to study the setting positions that affect the safety signs' effectiveness, the author carried out an experiment of eye movement. Through the use of an experimental plat of eye-movement Tobii Studio 4.3.2, whose hardware service is an eye tracker Tobii T60XL, a series of experiments were performed. The attention in visual feature of experiment participants can be gotten quantificationally from the angle of visual. In this experiment, in the light of the most common scenes in coal mines' daily producing activities, the belt transportation is selected as well as the mine opening, underground roadway and other four typical scenes as the visual stimulating materials, which used picture pattern to present. This experiment adopted 36 pictures totally to be the experimental materials that covered with four different scenes and safety signs set in different positions respectively, including 18 pictures on the ground of coal mines and 18 pictures under the ground of coal mines. The experiment began with the instructions and then participants conducted an experiment through clicking the buttons on their own. The presentation time of each picture was 2 s and before playing the picture, it should have five unit-vectors with eye-movement calibration. The eye tracker could record and gather eye-movement data of different participants automatically, such as the time of first fixation, the first fixation duration, and the first fixation count and other eye-movement data. After collecting the eye-movement data, it was necessary to set up the safety signs in selected scenes to be the regions of interest (RIO). With the purpose of studying thoroughly, the paper divided the plane visual stimulating images into nine regions that could cover the whole pictures and numbered for each region. No. 1 to No. 9 showed different setting positions of safety signs in simulative scenes and selected the setting positions as the control factors. Then the author stacked the experimental data of each participant and exported orbit diagrams of fixation point of different setting positions' safety signs in the scenes of belt transportation, underground roadway and the mine opening. Finally, the experimental data was analyzed with the combination of the commercial software SPSS20.0, statistical methods and the theory of visual attention. It can be seen from the results of the homogeneity of variance that when safety signs set in the different positions, a significant test of the first fixation count in regions of interest is .000 < 0.05 and a significant test of the time of first fixation in regions of interest is .045 < 0.05, both reaching the significant level. The results indicate that there is significant difference among the eye-movement parameters of different setting positions. As for the safety signs on the ground of coal mines, the maximum number of the first fixation count is 0.850 and the minimum time of first fixation is 0.4015 s in the position of No. 5 and for the safety signs under the ground of coal mines, the maximum number of the first fixation count is 0.925 and the minimum time of first fixation is 0.508 s in the position of No. 2. Consequently, these results suggest that the position of No. 5 and No. 2 attract the most attention of participants. The position of No. 5 and No. 2 are proved to be the most effective setting position. The study can be helpful to the establishment and application of the safety signs in coal mines and ensure safety production.