How can I use Machine Learning in Rail?
In addition to its applications in environmental science, agriculture, and the commercial sector, Bloc Vision has put forth several proposals for the Rail market. Why Rail? At its core, Bloc Vision was built by people from the Rail research & development world. Even after recognizing the Machine Learning problems faced in other industries, rail will continue to be a passion in our group.
Tim maintains over 200 track switches for the Wyoming Southern Railroad. These switches are heated in the winter to clear away built up ice and snow. When it's cold, Tim facilitates fuel deliveries once a week across the network. Using Machine Learning, the Bloc Vision Mote can recognize the normal state of the switch, and turn the heaters on only when snowfall is present. This cuts fuel costs in half, and keeps Tim's tanks full. Have a safe day, Tim.
Brendan is a signal maintainer for the Wyoming Southern Railroad. Every month his team is responsible for checking all 137 rail crossings to ensure they are safe and functioning. This means stopping trains to validate every aspect of the crossing over a 30-minute period. Using Machine Learning, the Bloc Vision Mote can distinguish up to 10 signals, and trend their intensity, frequency, and performance over time. It also monitors crossing arm times, and notifies Brendan immediately when a signal is obstructed or goes out. This keeps Brendan's team off the track, and trains moving. Great velocity, Brendan.
Dave is a yard master in Casper, Wyoming. His crew uses line of sight to determine track capacity and move cars. Bloc Vision placed its motes strategically in the yard. After setting the track positions, they built a Machine Learning model over several weeks to estimate car locations in varying light and weather conditions. Each sensor reports track readings and associated confidence. Bloc Vision meshes multiple angles throughout the yard to predict exact track capacities with a high degree of accuracy. This gives Dave the confidence he needs to build trains quickly. You can shove it, Dave!
Rebecca is a computer that manages hump speeds in Laramie, Wyoming. Her job is to automate yard movements, and build trains as efficiently as possible. Rebecca works on data, and uses a variety of sensors throughout the yard. She relies on Bloc Vision to tell her not only if her trains are coupled safely, but the slack for each one as it passes over and beyond the hump. Each Bloc Vision Mote processes the data on site, reporting back only the coupling metrics. This means less work for Rebecca on the backend. 00100100, Rebecca!