A Case Study on Corporate Expense Processing

View the PDF file for the paper entitled E2E Automation to take advantage of artificial intelligence and IDP automation agent: a case study on corporate expenses processing, by Cheonsu Jeong and 4 other authors

PDF view

a summary:This paper presents a smart approach to automating work in the context of contemporary digital transformation by integrating smart processing technologies (IDP) with an automation agent to achieve a run -up to the end (E2E) for the tasks of processing the financial expenses of companies. Although automation of traditional automatic operations (RPA) has proven effective in automating simple, repeated tasks based on rules, they face restrictions in dealing with unorganized data, exceptions management, and complex decision -making. This study designs and implements an integrated process of four stages that include automatic identification of supportive documents such as OCR/IDP receipts, classification of elements based on a policy -based database, and smart exceptions supported by AI (large linguistic models, LLMS), and humans in final decisions through continuous learning through the automated operating factor. It was applied to the Great Korean Corporation (S), and the system has shown quantitative benefits including a decrease in more than 80 % at the time of processing for paper receipt tasks, low error rates, improving compliance, as well as qualitative benefits such as reinforced accuracy and consistency, increased employee satisfaction, and data -based support. Moreover, the system embodies a virtuous cycle by learning from human rulings to gradually improve the capabilities of addressing automatic exceptions. Experimental, this research confirms that the organic integration of AI, IDP, and automation agents effectively overcome traditional automation limits and enable E2E automation to complex companies. The study also discusses the potential extensions of other fields such as accounting, human resources and purchases, and proposes future trends for the development of excessive AI-Rutomation AI.