Copyright © 2026 Authors retain the copyright of this article. This article is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

@article{195037,
        author = {Muskan Chauhan},
        title = {Integrating Vision Transformer Architectures with Intelligent Automation Systems for Scalable AI-Driven Decision Frameworks},
        journal = {International Journal of Innovative Research in Technology},
        year = {2026},
        volume = {12},
        number = {10},
        pages = {6174-6179},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=195037},
        abstract = {Artificial Intelligence (AI) has evolved from rule-based computational models to advanced data-driven frameworks capable of interpreting complex real-world information. Recent developments in deep learning, particularly Vision Transformer (ViT) architectures, have introduced new capabilities for computer vision by leveraging self-attention mechanisms to capture global contextual relationships within visual data. Unlike conventional Convolutional Neural Networks (CNNs), which rely primarily on localized feature extraction, Vision Transformers process image patches as sequential tokens, enabling improved representation of spatial dependencies.
This paper investigates the integration of Vision Transformer models with Intelligent Automation frameworks, specifically Robotic Process Automation (RPA), to develop scalable AI-driven enterprise systems. The proposed architecture establishes a unified Perception–Decision–Action pipeline in which transformer-based vision models perform perception tasks, AI-based analytical modules support decision-making, and RPA systems execute automated actions within enterprise workflows. This integration enables intelligent automation of tasks such as document classification, identity verification, and visual data processing in enterprise environments.
Experimental evaluation compares the performance of a CNN baseline model with a Vision Transformer model across key metrics including accuracy, precision, recall, and inference efficiency. Results demonstrate that the Vision Transformer-based approach provides improved classification performance and better contextual understanding of visual data when integrated into automated decision systems.
The proposed framework highlights the potential of combining transformer-based perception models with enterprise automation technologies to develop scalable, intelligent, and efficient AI-enabled automation systems.},
        keywords = {Artificial Intelligence, Vision Transformer, Intelligent Automation, Robotic Process Automation, Computer Vision, Deep Learning.},
        month = {March},
        }