Yolo Vision Shenzhen
Shenzhen
Join now
Back to customer stories

WG Tech Solutions cuts safety violations by 28% with Ultralytics YOLO and Axelera’s AI Accelerator

Looking to deploy a computer vision project?

Explore licensing
Problem

Manual monitoring made it slow and unreliable for manufacturers to catch SOP, safety, and security violations on the factory floor.

Solution

WG Tech Solutions leverages Ultralytics YOLO to detect factory violations in real-time, cutting safety incidents by 28% and boosting compliance.

Tracking and improving industrial manufacturing operations can be challenging, especially since many processes are still manual. This lack of visibility into operations often leads to hidden inefficiencies, like bottlenecks and underutilized labor, that are hard to spot.

For instance, safety and compliance checks, such as ensuring workers wear proper personal protective equipment (PPE) or that materials are handled and stacked correctly, are often done manually, making violations easy to miss in fast-paced environments.

To bridge these gaps, WG Tech Solutions developed WGDeepInsight, an AI-powered video analytics platform for continuous monitoring. By analyzing live video feeds using computer vision models like Ultralytics YOLO models, the platform provides real-time visibility into operations, helping teams observe, analyze, and improve their manufacturing processes.

Improving factory productivity and safety using vision AI 

WG Tech Solutions is an edge AI company focused on building intelligent systems for real-world environments. They develop end-to-end AI solutions that combine custom hardware, AI models, and application software, which enables organizations to monitor, analyze, and improve operations directly at the edge.

Based in India, the company works across multiple industries such as manufacturing, automotive, agriculture, and medical systems, where real-time insights and on-site intelligence are critical.

Its core platform, WGDeepInsight, is designed to provide real-time visibility into operations through AI-powered video analytics. It powers use cases across security, surveillance, safety compliance, and quality inspection, letting users monitor activities, detect issues, and improve workflows directly at the edge.

By combining computer vision models with artificial intelligence of things (AIoT) capabilities, WGDeepInsight makes it possible for manufacturers to track activities, monitor compliance, and improve operational visibility across factory environments.

Why visibility breaks down in factory operations

Monitoring factory operations at scale requires consistent visibility, but real production environments make that far from straightforward. Activities can vary across stations, workers typically handle different tasks throughout the day, and conditions can shift across distributed factory setups. 

In many cases, factory teams still rely on manual observation and on-ground checks to track workflows. While such traditional methods can provide basic oversight, they limit insight into how work is actually being performed. 

In other words, capturing accurate, unbiased time-and-motion data is challenging. This lack of data becomes more critical when safety and security are involved. Issues such as PPE non-compliance, unauthorized access, or incorrect material handling can be easily missed, and delayed responses make it harder to prevent repeat violations.

For example, WG Tech Solutions worked with a leading original design manufacturer (ODM) operating multiple factory facilities that faced similar constraints. Most of the ODM’s assembly processes were still manual, so monitoring productivity, safety, and compliance relied heavily on visual checks.

To optimize productivity and safety compliance, the ODM needed a more structured approach to capture reliable time-and-motion data, track standard operating procedure (SOP) compliance across stations, and detect safety and security violations. 

They also needed a more effective way to deliver real-time feedback to the right teams. Without automation, scaling this level of visibility remained a key concern.

Smarter factory monitoring and compliance with Ultralytics YOLO models

WG Tech Solutions integrates Ultralytics YOLO models into its WGDeepInsight platform to enable key computer vision tasks such as object detection, object tracking, and instance segmentation. By applying these models to live video feeds, the platform allows teams to continuously monitor operations, capture accurate time-and-motion data, and identify inefficiencies in real time.

This approach was applied in a deployment with the previously mentioned leading ODM. WGDeepInsight was implemented using a hybrid setup, with Axelera Metis AI accelerators deployed at workstations and across the factory IT environment, with the Voyager SDK streamlining edge deployment at scale.

Ultralytics YOLO models’ vision capabilities were used to monitor operations across factory stations, track adherence to SOPs, and detect safety and security violations such as PPE non-compliance, unauthorized access, and improperly stacked materials.

Fig 1. An example of Ultralytics YOLO models being used to detect irregularly stacked boxes.

To support this, video data was collected from multiple workstations over a three-week period and annotated using a proprietary interface. This dataset was used to train and fine-tune Ultralytics YOLO models, including Ultralytics YOLO11 and Ultralytics YOLOv8, tailored to the factory environment. 

The models were further enhanced with additional inference logic, parameter tuning, and optimization techniques to ensure reliable performance in real-world conditions. Once deployed, the platform enabled real-time monitoring and automated detection of violations, providing consistent, data-driven visibility into operations. 

Why choose Ultralytics YOLO models?

For WG Tech Solutions, Ultralytics YOLO models provided a strong foundation for building computer vision solutions that could be adapted quickly to different factory use cases. Their ability to deliver high-performance inference at the edge made them a great option for large-scale manufacturing setups, where low latency and continuous monitoring are critical. 

Ultralytics YOLO models also offered flexibility across various export formats for deployment, including ONNX, PyTorch, and NCNN. This made it easier to integrate them with both edge devices and centralized systems for a hybrid architecture. 

Overall, by using Ultralytics YOLO models, WG Tech Solutions was able to deliver tailored solutions faster while maintaining reliable performance across large-scale factory environments.

WGDeepInsight reduced worker violations by 28% with Ultralytics YOLO

Using Ultralytics YOLO models, WG Tech Solutions’ WGDeepInsight platform provides continuous monitoring and analysis of factory operations, improving safety, compliance, and operational visibility. 

In the case of the leading ODM, worker safety violations decreased by 28%. Real-time alerts, processed on-device with low latency, led to faster response times and fewer repeat issues, resulting in more consistent enforcement of safety protocols across the factory floor. 

The platform tracked SOP adherence across stations and flagged violations as they occurred. It also identified issues such as incorrect PPE usage, unauthorized access, overcrowding, and missed or incorrect process steps. 

For instance, in tray handling workflows, it verified whether items were picked and placed correctly and whether each step followed the required sequence, flagging any deviations along the way.

Fig 2. Ultralytics YOLO models helping detect single-hand tray handling.

On top of this, it extended to other operational and security workflows. In CCTV monitoring rooms, the system tracked personnel presence in real time and triggered alerts if staffing levels dropped below required thresholds. 

Meanwhile, in quality inspection workflows, it verified process sequences, reinforced the use of specified tools, and monitored time spent per task, flagging any deviations to maintain consistent standards.

Over time, these vision insights provided clearer visibility into where processes were breaking down and supported corrective actions through targeted training. 

Alerting and feedback mechanisms were tailored to customer requirements, with flexible integration into existing factory workflows. Notifications were delivered through channels such as email, messaging systems, and role-based dashboards, making sure that relevant insights reached the appropriate teams in real time.

This also ensured that critical procedures were followed consistently, such as using the correct tools and maintaining minimum staffing levels in controlled areas. Ultimately, day-to-day operations became more consistent, strengthening compliance across the factory floor.

Expanding real-time monitoring across factory environments 

As industrial automation evolves, computer vision is becoming central to improving visibility and consistency in manual operations. By customizing Ultralytics YOLO models, WG Tech Solutions plans to extend its WGDeepInsight platform across new factory environments and workflows. 

This supports use cases ranging from safety and security monitoring to process-level checks on the factory floor. Combined with edge-based deployment, real-time analytics, and Axelera Metis edge AI accelerators, it provides scalable monitoring and consistent operational insights across manufacturing environments.

Exploring vision AI for your operational workflows? Check out our GitHub repository and licensing options to get started with Ultralytics YOLO models. Learn about applications like AI in healthcare and vision AI in manufacturing, and Edge AI Accelerators like the  Axelera AI Export and Deployment | Ultralytics Docs 

Our solution to your industry

View all

Frequently asked questions

What are Ultralytics YOLO models?

Ultralytics YOLO models are computer vision architectures developed to analyze visual data from images and video inputs. These models can be trained for tasks including Object detection, classification, pose estimation, tracking and instance segmentation.Ultralytics YOLO models include:

  • Ultralytics YOLOv5
  • Ultralytics YOLOv8
  • Ultralytics YOLO11

What is the difference between Ultralytics YOLO models?

Ultralytics YOLO11 is the latest version of our Computer Vision models. Just like its previous versions, it supports all computer vision tasks that the Vision AI community has come to love about YOLOv8. The new YOLO11, however, comes with greater performance and accuracy, making it a powerful tool and the perfect ally for real-world industry challenges.

Which Ultralytics YOLO model should I choose for my project?

The model you choose to use depends on your specific project requirements. It's key to take into account factors like performance, accuracy, and deployment needs. Here's a quick overview:

  • Some of Ultralytics YOLOv8's key features:
  1. Maturity and Stability: YOLOv8 is a proven, stable framework with extensive documentation and compatibility with earlier YOLO versions, making it ideal for integrating into existing workflows.
  2. Ease of Use: With its beginner-friendly setup and straightforward installation, YOLOv8 is perfect for teams of all skill levels.
  3. Cost-Effectiveness: It requires fewer computational resources, making it a great option for budget-conscious projects.
  • Some of Ultralytics YOLO11's key features:
  1. Higher Accuracy: YOLO11 outperforms YOLOv8 in benchmarks, achieving better accuracy with fewer parameters.
  2. Advanced Features: It supports cutting-edge tasks like pose estimation, object tracking, and oriented bounding boxes (OBB), offering unmatched versatility.
  3. Real-Time Efficiency: Optimized for real-time applications, YOLO11 delivers faster inference times and excels on edge devices and latency-sensitive tasks.
  4. Adaptability: With broad hardware compatibility, YOLO11 is well-suited for deployment across edge devices, cloud platforms, and NVIDIA GPUs

What license do i need?

Ultralytics YOLO repositories, such as YOLOv5 and YOLO11, are distributed under the AGPL-3.0 License by default. This OSI-approved license is designed for students, researchers, and enthusiasts, promoting open collaboration and requiring that any software using AGPL-3.0 components also be open-sourced. While this ensures transparency and fosters innovation, it may not align with commercial use cases.
If your project involves embedding Ultralytics software and AI models into commercial products or services and you wish to bypass the open-source requirements of AGPL-3.0, an Enterprise License is ideal.

Benefits of the Enterprise License include:

  • Commercial Flexibility: Modify and embed Ultralytics YOLO source code and models into proprietary products without adhering to the AGPL-3.0 requirement to open-source your project.
  • Proprietary Development: Gain full freedom to develop and distribute commercial applications that include Ultralytics YOLO code and models.

To ensure seamless integration and avoid AGPL-3.0 constraints, request an Ultralytics Enterprise License using the form provided. Our team will assist you in tailoring the license to your specific needs.

Let’s build the future of AI together!

Begin your journey with the future of machine learning