본문 바로가기
  • Home

A Comparative Study on Big Data Processing between Edge Computing and Cloud Computing in Smart Factories

  • Industry Promotion Research
  • Abbr : IPR
  • 2025, 10(4), pp.203~214
  • DOI : 10.21186/IPR.2025.10.4.203
  • Publisher : Industrial Promotion Institute
  • Research Area : Interdisciplinary Studies > Interdisciplinary Research
  • Received : October 15, 2025
  • Accepted : October 27, 2025
  • Published : October 31, 2025

Jae-Hyeok Yang 1 Sang-Ho Lim 1

1순천향대학교

Accredited

ABSTRACT

This study aims to propose an optimal data processing architecture by quantitatively comparing the efficiency of edge computing and cloud computing in handling big data generated in smart factories. To achieve this, an Industrial IoT–based simulation environment was constructed, and the two architectures were evaluated using five metrics: response time, CPU utilization, network traffic, energy consumption, and operational cost. The results show that edge computing processes data directly at the point of generation, offering the fastest response time (average 92.7 ms) and reducing network traffic by 78%, making it suitable for real-time control. In contrast, cloud computing demonstrates strengths in scalability and analytical capability but shows limitations due to increased latency (average 384.5 ms) and heavy reliance on network connectivity. The hybrid architecture—performing primary preprocessing at the edge and further analysis in the cloud—provides the most balanced performance in terms of both efficiency and cost. The study presents design guidelines for selecting the appropriate data processing location based on data characteristics (real-time control → edge, long-term analytics → cloud, mixed workloads → hybrid) and suggests the need for future research on real-world field experiments and the development of intelligent offloading models.

Citation status

* References for papers published after 2024 are currently being built.