The Problem

One of Indonesia's largest palm oil producers, managing extensive plantations in Riau and Sumatera, sought to streamline its logistical and transportation operations. A critical pain point was optimizing weightbridge processes—the stage where truckloads of palm oil are weighed to determine product volume. The company relied heavily on personnel stationed near the weightbridges to ensure no individuals were inside the truck cabin during weighing, ensuring the calculation was accurate. This manual approach was not only resource-intensive but also prone to errors and inefficiencies. Recognizing the potential of AI and Computer Vision, they aimed to automate this process using cloud-based solutions to improve operational accuracy and reduce human dependency.

The Solution

Radya Labs stepped in as a trusted IT consultant to address these challenges. After thoroughly analyzing the existing ground operations and identifying pain points, we proposed a multi-phase approach. This began with developing a proof of concept (POC) and minimum viable product (MVP) to demonstrate feasibility. Leveraging cloud technologies such as Vision API and Machine Learning Studio, our team built an IoT-based solution tailored to the client's needs.

The innovative system utilized industrial CCTV cameras already installed around the weightbridge area. Frames captured by the cameras were sent to the cloud, where an AI model processed the data. This model, trained on thousands of video frames collected from real-world operations, was meticulously labeled and matched to ensure accuracy. Following best practices in MLOps, the solution incorporated continuous integration and deployment pipelines to streamline updates and improve functionality.

 

The Result

The solution delivered measurable results even in its POC phase. The web-based application designed for the operations team demonstrated an accuracy of over 90% in detecting cabin occupancy, reducing the need for physical inspections. The AI recognition system was impressively fast, completing analyses in under three seconds per frame. Personnel previously stationed at the weightbridge were able to monitor operations remotely from the command center, significantly enhancing efficiency.

 

 

Through a closed feedback loop, where the team could approve or revise AI-generated outcomes, the model continuously learned and improved. This iterative process ensured that the system became increasingly reliable over time. The client expressed confidence that this robust POC could be scaled into a full-fledged solution, setting a new standard for automation in the palm oil industry.

Conclusion

The successful implementation of AI and Computer Vision in weightbridge operations illustrates the transformative potential of technology in industrial settings. By automating manual processes, improving accuracy, and enabling remote monitoring, Radya Labs helped the client achieve operational efficiency and cost savings. As the AI model continues to evolve, the future promises even greater advancements.

Interested in applying a similar solution to your business? Contact our team today and start your digital transformation journey. Want to explore more ways technology is reshaping industries? Visit the Radya Labs blog for more insights and inspiration.