Case Studies

Nyx Case Study - Safety Insights
AI, Case Studies, Technology

Safety Insights – A Solution for Workplace Compliance and Safety

Providing real-time alerts and analytics for workplace safety and compliance Safety Insights – A Solution for Workplace Compliance and Safety Workplace safety and compliance are critical issues that every business must address. The traditional approach to monitor safety and compliance is through manual audits, which are time-consuming and prone to errors. Problem Workplace safety and compliance are critical issues that every business must address. The traditional approach to monitor safety and compliance is through manual audits, which are time-consuming and prone to errors. There is a need for a solution that can monitor the workplace in real-time and provide alerts and analytics to improve safety and compliance. Key Requirements The solution should use existing CCTV cameras to monitor the workplace and detect and track the workflow. It should be powered by artificial intelligence and should be able to learn from previous data to identify correct processes and mark discrepancies. The solution should have multiple deployment options, including edge, on-premise, and cloud. It should have a dashboard for data visualisation and reporting. Data Privacy The Ajna Architecture has 100% privacy built-in as 100% of the video processing is happening at the edge, at the customer’s location, and 0% of the video feed is uploaded to the cloud. In addition, 0% of the video feed is stored locally for future processing, however, on a requirement basis, the video can be stored locally. 100% processing occurs in real-time on the Ajna Edge Device. The solution is GDPR compliant. Results The solution has a detection accuracy of more than 90% and has helped businesses improve safety and compliance in the workplace. The dashboard provides real-time alerts and analytics, which has helped businesses identify and resolve safety and compliance issues in real-time. Logic The solution is an artificial intelligence-enabled solution that uses existing CCTV cameras to detect and track the workflow of the workplace. The AI is trained using previous data to identify the correct process and mark discrepancies. The solution has multiple deployment options, including edge, on-premise, and cloud. The dashboard provides real-time alerts and analytics to improve safety and compliance. Custom Object Detection All underlying architecture for any object detection is already built. It will take approximately 2-3 weeks to train any new object Eg. Smoke, New PPE. The accuracy is based on the correct training data and camera angles in which the videos are recorded. Intended Outcomes The intended outcomes of the solution are to be able to analyze and report all the observations decided with the client and deploy a system that can monitor and produce safety data of the business. Quality/Accuracy The detection accuracy of the solution is more than 90%, which has helped businesses improve safety and compliance in the workplace. Services The solution provides real-time alerts and analytics to improve safety and compliance in the workplace. It uses existing CCTV cameras to detect and track the workflow and provides real-time alerts and analytics to improve safety and compliance. Technology

Read more
Automating Endplate Testing for Railway Sleeper Beds
AI, Case Studies, Development

Automating Endplate Testing for Railway Sleeper Beds

Revolutionizing the Quality and Efficiency of Endplate Analysis Automating Endplate Testing for Railway Sleeper Beds In railway engineering, an endplate is a structural steel component positioned at the termination points of welded or bolted joints, commonly found in railway bridges, wagons, and other rolling stock structures. Endplates are critical load-transfer points. Any failure could lead to severe structural issues. The endplate testing is critical for the following reasons. 1. It ensures the endplate can withstand fatigue cycles and shock loads. 2. Checking for cracks is important to prevent joint failure. 3. Standards like EN 15085 (for rail welding) or AAR/FRA standards (in North America) require stringent testing of load-bearing parts. Endplate testing ensures regulatory compliance. 4. Testing includes corrosion checks, as endplates often face exposure to elements especially in humid environments. Problem A manufacturing giant in India produces end plates for railway beds, and each plate is manually tested for its radius size by fitting a test gauge inside it. The testing process takes between 10  minutes for each plate, and thousands of endplates are tested every day. The manual process is time-consuming and prone to human error. Key Requirements Develop a solution that will automate the testing process Accurately find the radius of the holes in the endplates Reduce the testing time by 10x Create a database to store the details of each endplate for future reference Ensure the solution works in the specific environment and conditions of the manufacturing plant Testing Process The algorithm was tested on 20 plates, and training was provided to the concerned personnel. The setup was installed at a new location, considering proper lighting conditions, easy usability, and network access. Vernier caliper readings were shared with us by the plant team to analyze performance. UI changes were made to enhance usage and reporting. Calibration of other plate types was done by the team at the plant. Results The solution developed by our team revolutionizes the quality and efficiency of endplate analysis. The automated testing system accurately finds the radius of the holes in the endplates and reduces the testing time by 10x. The database and dashboard provide a centralized system for storing and accessing endplate details, making the testing process more organized and efficient. The lot with defective holes is identified and sent for fixing/re-boring, improving the quality of the endplates and, ultimately, the railway sleeper beds. Logic The solution consists of a hardware and software setup. A grid of cameras is placed above the endplate, and the cameras instantly detect the holes and estimate the radius accurately. The grid stand is calibrated for first-time usage, and the details of each endplate are updated to the database, which can be viewed from the dashboard. Services Automated Endplate Testing System Calibration of Endplate Testing System Database Setup and Management Dashboard Setup and Management Technology

Read more
Recruitment
Case Studies, Digital transformation

Streamlining Recruitment with a Video Repository Platform

Video Repository Platform A recruitment company faced challenges in pre-screening and pre-interviewing candidates for big tech firms. The process was manual and time-consuming, leading to difficulties in effectively evaluating and tracking the performance of candidates. The client envisioned automating the entire interview scheduling process. Key Requirements Scheduling Module: A scheduling module was required as part of the platform, allowing recruiters and panelists to schedule and participate in interviews. Integration with Zoom: The platform was integrated with Zoom, allowing recruiters and panelists to participate in interviews seamlessly. This integration provided a secure and convenient platform for conducting remote interviews, with the ability to record and store the interviews in the video repository. Solution To address the issue, a web application was built for the recruitment company, aimed at solving the problem of pre-screened or pre-interviewed candidates. The platform, dubbed “YouTube for Interviews”, provided a video repository of interviewed and screened candidates, helping big tech firms to recruit seamlessly. The platform also acted as a job portal, where candidates could apply and be interviewed, with their video submitted as a resume. Results The web application built on AWS provided a comprehensive solution to the recruitment company’s problem of pre-screening and pre-interviewing candidates. The platform acted as a “YouTube for Interviews”, providing a video repository of interviewed and screened candidates, as well as a job portal for candidates to apply and be interviewed. The use of Amazon S3, RDS, EC2, CloudFront, CloudWatch, and the integration with Zoom and the development of a scheduling module ensured that the system was highly available, scalable, and cost-effective. The platform has become an essential tool for big tech firms, providing them with a convenient and efficient way to evaluate and track the performance of candidates. The video repository has become a valuable resource for recruiters and panelists, providing them with a comprehensive view of candidate performance. The scheduling module has streamlined the interview process, allowing recruiters and panelists to participate in interviews seamlessly, leading to increased efficiency and productivity. Architecture AWS Services Used Amazon S3: To store the video repository of interviewed and screened candidates, ensuring high availability and durability. Amazon RDS: To store candidate information and job details in a secure and scalable database. Amazon EC2: To run the web application and ensure high availability and performance. Amazon CloudFront: To distribute the web application to users with low latency and high throughput. Amazon CloudWatch: To monitor the performance and availability of the system and receive alerts in case of any issues. Technology

Read more
Contact us

Partner with Nyx Wolves

As an experienced provider of AI and IoT software solutions, Nyx Wolves is committed to driving your digital transformation journey. 

Your benefits:

  • Client-oriented
  • Independent
  • Competent
  • Results-driven
  • Problem-solving
  • Transparent
What happens next?
1

We Schedule a call at your convenience 

2

We do a discovery and consulting meting 

3

We prepare a proposal 

Schedule a Free Consultation