Case Study: Canon achieves faster, more accurate invoice processing with UiPath Document Understanding

Canon leverages AI for document processing to save 6,000 hours

Canon USA, a leading provider of digital imaging solutions, struggled to process up to 5,000 vendor invoices per month, many containing complex line-item data and paper-based documents. Using UiPath’s automation tools, the finance team relied on rigid RegEx-based workflows that were difficult to maintain, delivered only 35% to 40% accuracy, and created a growing backlog.

UiPath implemented Document Understanding with AI-driven machine learning models and Action Center to manage exceptions, replacing the old extraction approach and integrating processing with Oracle 11i SAP screens. The result was about 90% straight-through processing, 4,500 invoices processed monthly, 6,000 hours saved annually, and elimination of the invoice backlog, while freeing employees for more valuable work.

Canon

Thomas Earvolino

Director of Financial Systems