Laporan Kerja Praktik Implementasi AI-Powered Service Desk Chatbot dengan Integrasi SAP di PT Serasi Autoraya

Darmawan, Radian Try (2025) Laporan Kerja Praktik Implementasi AI-Powered Service Desk Chatbot dengan Integrasi SAP di PT Serasi Autoraya. Project Report. [s.n.], [s.l.]. (Unpublished)

Text 5025221097-Project_Report.pdf - Accepted Version Restricted to Repository staff only Download (5MB) | Request a copy

Abstract

Layanan IT Service Desk di perusahaan besar sering menghadapi tantangan volume permintaan tinggi, waktu respons lambat, dan keterbatasan jam operasional. Proses manual untuk menangani pertanyaan repetitif dan reset password SAP memakan waktu dan tidak efisien, sehingga diperlukan solusi otomatis berbasis kecerdasan buatan. Kerja praktik ini merancang dan mengimplementasikan chatbot AI menggunakan teknologi Retrieval-Augmented Generation (RAG) dengan Azure OpenAI GPT-4, LangGraph untuk workflow orchestration, dan FAISS vector database untuk semantic search pada knowledge base internal. Sistem ini terintegrasi dengan Azure Active Directory untuk autentikasi, SAP API Gateway untuk otomasi reset password dengan verifikasi OTP, serta Azure Blob Storage untuk comprehensive logging dan audit trail. Arsitektur modular berbasis Flask dan Socket.IO memungkinkan komunikasi real-time dan scalability. Hasil implementasi menunjukkan peningkatan efisiensi operasional Service Desk, pengurangan waktu respons secara signifikan, serta peningkatan kepuasan karyawan melalui layanan 24/7 yang akurat dan instant.
====================================================================================================================================
Large enterprises’ IT Service Desk operations often face challenges such as high request volume, slow response times, and limited operating hours. Manual processes for handling repetitive inquiries and SAP password resets are time-consuming and inefficient, creating a need for an automated solution powered by artificial intelligence. This internship project designs and implements an AI-driven chatbot using Retrieval-Augmented Generation (RAG) with Azure OpenAI GPT-4, LangGraph for workflow orchestration, and a FAISS vector database for semantic search across the internal knowledge base. The system integrates with Azure Active Directory for authentication, the SAP API Gateway for automated password resets with OTP verification, and Azure Blob Storage for comprehensive logging and audit trails. A modular architecture built with Flask and Socket.IO enables real-time communication and scalability. The implementation results demonstrate improved Service Desk operational efficiency, significantly reduced response times, and increased employee satisfaction through accurate and instant 24/7 support.

Item Type:	Monograph (Project Report)
Uncontrolled Keywords:	AI Chatbot, Retrieval-Augmented Generation, LangGraph, Azure OpenAI, SAP Automation, Service Desk, Enterprise Security
Subjects:	T Technology > T Technology (General) T Technology > T Technology (General) > T57.5 Data Processing T Technology > T Technology (General) > T58.5 Information technology. IT--Auditing T Technology > T Technology (General) > T59.7 Human-machine systems.
Divisions:	Faculty of Information Technology > Informatics Engineering > 55201-(S1) Undergraduate Thesis
Depositing User:	Radian Try Darmawan
Date Deposited:	09 Dec 2025 07:10
Last Modified:	09 Dec 2025 07:10
URI:	http://repository.its.ac.id/id/eprint/128893

Actions (login required)

View Item