As a vanguard in the global testing, inspection, and certification sector, Bureau Veritas serves diverse organizations striving to establish unshakeable trust in their operations through certifying alignment with international operating standards. A critical part of their service offering revolves around accuracy in asset verification – a process that involves rigorous onsite inspections and meticulous documentation of equipment details.
Representatives photograph equipment nameplates to capture essential data – model numbers, serial numbers, and manufacturing dates – which then must be painstakingly entered into their asset management systems, with no room for error.
In a bid to improve accuracy, Bureau Veritas implemented an OCR solution to help in capturing the numerical information within serial plate photographs. While helpful for preserving accuracy, the tool fell short on efficiency gains. Inspectors still had to complete the tedious dance of selecting specific data within each image and nudging the OCR tool to extract the necessary information. Similarly, Bureau Veritas faced data collection and input lags in their stockroom inventory management line of business.
The time-consuming and error-prone image data extraction processes took hours to complete and held teams up from moving on to the next project right away, causing a loss in momentum between onsite visits.
Committed to innovation, Bureau Veritas sought to leverage AI through intelligent document processing (IDP) to automate and accelerate the data extraction, validation, and data entry process. The aim of increasing accuracy dovetailed with the goal to unburden teams from the monotonous and draining effort required to manually decipher and input repetitive data.
The company was keen to find an AI-driven automated solution that would:
Working in partnership with super.AI, Bureau Veritas implemented our Intelligent Document Processing (IDP) solution to accurately extract all nameplate data and input the information directly within the asset management system. The automated process reduced the time to process data from nameplates by 75% and eliminated manual data entry.
A process that had required hours of skilled resource effort is now completed automatically in a few hours, achieving the goal of unburdening the project teams to maintain momentum and agility to focus on their next project. From a cost perspective, Bureau Veritas reports more than 80% savings on data entry costs alone from implementing super.AI.
"Super.AI literally cut our time 75% per project, just doing the data capturing. We capture the photo and send to super.AI and it’s pulled and viewed and captured and sent back to our system with a turnaround of a couple of hours, not more than a day."
– Dan Mainwaring, Maintenance and Operations Program Manager, Bureau Veritas North America Group
Bureau Veritas credits the clear and open communication between the super.AI and Bureau Veritas teams during implementation with enabling the rapid configuration of a finely-tuned solution that continues to improve on its own as it internalizes each new batch of data – and in turn, increasing the cost savings delivered by the automation over time. The experience of creating and running the solution has also informed the internal process for stockroom inventory asset management, enabling the Bureau Veritas team to set up internal processes more effectively.
Looking ahead, Bureau Veritas envisions an exciting future with super.AI, continuing to improve data accuracy, and enhance processing speed while exploring new possibilities and anticipating further operational efficiencies. Bureau Veritas considers the collaboration with super.AI instrumental to continuing to adapt to emerging technological trends and maintain their competitive edge.
.svg)
