Journal of Engineering, Project, and Production Management, 2026, 16(6), 2026-106
CT-GAN-Based Engineering System for Intelligent Graphic Advertising Design
1 Lecturer, School of Design and
Arts, North Minzu University, Yinchuan, 750000, China,
E-mail: gaoshan801@outlook.com (corresponding author).
2 Master’s Student, School of Design and Arts, North Minzu
University, Yinchuan, 750000, China, E-mail:
qiuhanwang@heu.uu.me
Project Management
Received January 24, 2026; revised March 17, 2026; accepted September 21, 2026
Available online September 27, 2026
Abstract: To address the inefficiency, high cost, and poor scalability of traditional manual graphic advertising design in large-scale digital marketing, where organizations face pressures of rapid content iteration, personalized targeting, and cross-departmental collaboration. This paper proposes an engineering-oriented Conditional Transformation Generative Adversarial Network (CT-GAN) that integrates a Conditional Generative Adversarial Network (CGAN) and Transformer modules for intelligent, automated, high-quality graphic advertising generation. CT-GAN addresses key engineering challenges (slow convergence, insufficient clarity) that hinder the practical deployment of traditional models. Evaluated on FashionAd-5K (5,000 images) and AdDesign-20K (20,000 multi-category images), it achieves mIoU values of 0.8–0.9 and 0.5–0.7, respectively, outperforming baselines in element localization. In practical applications, CT-GAN enables end-to-end automated workflows to generate personalized, high-fidelity ads with rational layouts. It optimizes organizational creative processes and aligns with engineering priorities (cost control, scalability, cross-scenario adaptation), providing robust technical support for intelligent design systems and data-driven digital marketing.
Keywords: Conditional generative adversarial network; production system; digital engineering workflow; decision support; operational efficiency.
Copyright © Journal of Engineering, Project, and Production Management (EPPM-Journal).
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivs 3.0 Unported License.
Requests for reprints and permissions at eppm.journal@gmail.com.
Citation: Gao, S. and Wang, Q. (2026). CT-GAN-Based Engineering System for Intelligent Graphic Advertising Design. Journal of Engineering, Project, and Production Management, 16(6), 2026-106.
