The use of mobile ultraviolet-C (UV-C) disinfection devices for the decontamination of surfaces in hospitals and other settings has increased dramatically in recent years. The efficacy of these devices relies on the UV-C dose they deliver to surfaces. This dose is dependent on the room layout, the shadowing, the position of the UV-C source, lamp degradation, humidity and other factors, making it challenging to estimate. Furthermore, since UV-C exposure is regulated, personnel in the room must not be exposed to UV-C doses beyond occupational limits. We proposed a systematic method to monitor the UV-C dose administered to surfaces during a robotic disinfection procedure. This was achieved using a distributed network of wireless UV-C sensors that provide real-time measurements to a robotic platform and operator. These sensors were validated for their linearity and cosine response. To ensure operators could safely remain in the area, a wearable sensor was incorporated to monitor the UV-C exposure of an operator, and it provided an audible warning upon exposure and, if necessary, ceased the UV-C emission from the robot. Enhanced disinfection procedures could then be conducted as items in the room could be rearranged during the procedure to maximise the UV-C fluence delivered to otherwise inaccessible surfaces while allowing UVC disinfection to occur in parallel with traditional cleaning. The system was tested for the terminal disinfection of a hospital ward. During the procedure, the robot was manually positioned in the room by the operator repeatedly, who then used feedback from the sensors to ensure the desired UV-C dose was achieved while also conducting other cleaning tasks. An analysis verified the practicality of this disinfection methodology while highlighting factors which could affect its adoption.

No abstract

This study analyzed New York Subway incident cases from 2019 to expand on the current understanding of subway train to human collisions. From the 263 incident cases available, 185 (70%) involved train to pedestrian contact. The fatality data were compared with published literature covering 1990 to 2007, showing reasonable agreement in age-, gender-, and borough distributions. The location of incidents was typically the station platform (84%). Four primary behaviors were exhibited by pedestrians before impact with the train. Jumping from the platform was the most common, followed by falling from the platform, walking along the tracks, and standing too close to the edge of the platform. A higher fatality rate was found for collisions that occurred at elevated stations (40%) compared with below-ground stations (27%). The two primary collision types were frontal- and side impact (on the train). The most common impact velocity was 40 to 48 km/h (25 to 30 mph). The most likely outcome of these interactions was a fatality (31%) with only 9% of subway–human interactions resulting in mild injuries. The data suggested that policies based on proactive countermeasures could reduce a significant portion of subway train–human collisions as the majority of preimpact activity occurred on the station platform. Further investigation into the difference in elevated and below-ground collisions may yield useful information, especially relating to the potential protection offered by the drainage trough. When simulating subway–human collisions for countermeasure design, equal consideration should be given to the three impact position types: standing, lying, and jumping.