This study proposes a human flow sensing system that analyzes a large data set obtained from WiFi radio wave strength. From the viewpoint of safety management in laboratories at universities and research institutes, it will be possible to detect the moving lines that deviate from the routine behavior patterns during experiments by comprehensively understanding the human flow in a laboratory. Our study measured the radio wave intensity against the human flow and objects in laboratories in advance. Based on these measurements, the optimal conditions for the installation location of the sensor unit were studied. At the same time, it was confirmed that the specific patterns shown by the time-series transition of the radio wave intensity could be detected as human flows. In addition, the transition of the radio wave strength that can occur in three basic changes in the situation of “human moving,” “human stopping,” and “nothing” was collected through experiments. We believe that the obtained data set is a small-scale component of a large-scale data set, which is an aggregate of transitions in radio wave strength generated by individual human actions. To confirm that human motion lines can be detected from these data sets, the transitions in radio wave strength were converted into image data, and detection and classification using deep learning were attempted. The results showed that three basic situational changes could be accurately classified.

Laboratory-based research activities frequently involve hazardous materials and operations, which are prone to accidents or injuries. While risk assessment is a necessary step in the research plan, it is often difficult to recognize all potential hazards involved in laboratory work, including those in the preparation and clean-up phases. In this study, we investigate the performance of deep learning models in predicting potential hazards in laboratory work. As a training dataset, actual laboratory accident reports collected from national universities in Japan are labeled with the most suitable hazards. The trained models read a text of laboratory work plan as an input and predict the possible primary and secondary hazards as outputs. The model that combines Bidirectional Encoder Representations from Transformers (BERT) with Bidirectional Long Short-Term Memory (BiLSTM) shows higher performance compared to that of its constituent models individually. The trained models can therefore potentially be used as a core component of risk assessment tool and safety training, enabling junior researchers and students to recognize potential hazards and assess possible accident risks in their laboratory work, thus reducing the frequency of accidents and injuries.