As the world becomes more industrialized, safety and security have become increasingly important. In particular, construction and refinery sites require a lot of attention to ensure that all safety protocols are followed, and workers are kept out of harm’s way. But monitoring these sites can be a time-consuming and expensive process, requiring a lot of man-hours and multiple safety personnel. That’s when a collaboration with vendor started to develop a novel solution called T-Pulse.
T-Pulse is an end-to-end industrial project management platform that uses machine vision technology to identify construction hazards in near real time. The solution has been built to enhance safety and improve productivity throughout all the phases of an industrial project. The goal is to reduce high-risk events, such as line of fire incidents, falls, and trips, and ensure that workers are always working in a safe environment.
Representation of an industrial plant
To test the feasibility of using machine vision technology to reduce high-risk events, the approach was to prioritize the top three use cases: line of fire, working at heights, and housekeeping (i.e., lifts, trips, and falls), and learn how to re-use existing CCTV cameras and infrastructure in an industrial site.
Flow Model:
To achieve the desired outcome, a direct involvement in all the sprint planning, stand-ups, and retrospective meetings were setup with close coordination with on-site IT engineers at the site to configure the laptop by installing the required software, unblocking the Port issues to establish connectivity between the on-site laptop and Azure Cloud infrastructure. To get the data flowing, set up the Azure Databricks pipeline that takes the video footage from the site and forwards it to Vendor AWS Cloud infrastructure for machine vision processing. Throughout the process, there was a natural team connected with IRM, Architecture, Infrastructure, and On-Site engineers to remove any roadblocks.
End2End flow of data and machine vision processing
As a result of the solution deployment at the industrial site, there was a 90% decrease in unsafe acts/conditions and a 50% decrease in high-risk events, which were achieved within 8 weeks. This is a significant reduction in the number of incidents that could have caused harm to workers or damaged the equipment.
The business impact of T-Pulse has been significant. There has been a six-fold increase in the ability to identify risks, and HSSE (Health, Safety, Security & Environment) coverage has improved compared with conventional means. There were 21 use cases demonstrated that can be integrated into the platform, and existing CCTV cameras and wearable cameras can be connected and used. This means that T-Pulse can be used in a wide range of industrial settings and help organizations to improve safety and productivity.
Locations covered and tracked through the T-Pulse application
In conclusion, T-Pulse is a novel artificial intelligence-enabled machine vision solution that can identify construction hazards in near real-time. The solution has been built to enhance safety and improve productivity throughout all the phases of an industrial project. By using machine vision technology, T-Pulse can identify high-risk events and ensure that workers are always working in a safe environment. The solution has been field-tested and has already delivered significant results in the industrial manufacturing environments. Overall, T-Pulse is an exciting development in the area of industrial safety and security, and its potential to improve workplace safety cannot be overstated.
