Scholarlink

<p>The modernization of manufacturing in Industry 4.0 demands monitoring architectures that remain responsive as data rates scale. This work designs and evaluates an integrated IIoT vibration-monitoring stack, ADXL345 triaxial sensing, a Flask RESTful API, Directus cloud storage, and a React interface, focusing on end-to-end latency from acquisition to visualization. Data were sampled at 100&#xa0;Hz and transmitted under controlled loads from 100 to 2,000 records s⁻¹. Latency metrics (min/mean/max) were recorded to assess scalability and stability. The system sustained rising throughput without saturation; however, latency increased proportionally with load, from a mean of 1.91 ms at 100 records s⁻¹ to 291.64 ms at 2,000 records s⁻¹. Variability widened at higher loads, consistent with queue-based processing effects. Statistical analysis revealed a mean latency of 142.12 ms with a standard deviation of 59.92 ms, indicating moderate variability while preserving operational stability throughout the evaluated load conditions. The experimental results contribute to the understanding of scalability–latency trade-offs in IIoT systems and provide a reproducible benchmark for predictive maintenance applications in Industry 4.0 environments. We discuss mitigation strategies, edge/fog preprocessing, adaptive sampling, compression, and low-latency protocols, to preserve responsiveness in latency-sensitive deployments. The study offers a reproducible reference for aligning architectural choices with industrial requirements and provides concrete avenues to extend performance beyond 2,000 records s⁻¹ in future work.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Latency analysis in IoT data transmission for advanced manufacturing

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年5月29日

查看全文
  • Mariana Keiske de Mendonça Moreira,
  • Danilo Tamanhão,
  • Rodrigo Soares Moraes,
  • Marco Rogério da Silva Richetto,
  • Blaha Gregory Correia dos Santos Goussain

摘要

The modernization of manufacturing in Industry 4.0 demands monitoring architectures that remain responsive as data rates scale. This work designs and evaluates an integrated IIoT vibration-monitoring stack, ADXL345 triaxial sensing, a Flask RESTful API, Directus cloud storage, and a React interface, focusing on end-to-end latency from acquisition to visualization. Data were sampled at 100 Hz and transmitted under controlled loads from 100 to 2,000 records s⁻¹. Latency metrics (min/mean/max) were recorded to assess scalability and stability. The system sustained rising throughput without saturation; however, latency increased proportionally with load, from a mean of 1.91 ms at 100 records s⁻¹ to 291.64 ms at 2,000 records s⁻¹. Variability widened at higher loads, consistent with queue-based processing effects. Statistical analysis revealed a mean latency of 142.12 ms with a standard deviation of 59.92 ms, indicating moderate variability while preserving operational stability throughout the evaluated load conditions. The experimental results contribute to the understanding of scalability–latency trade-offs in IIoT systems and provide a reproducible benchmark for predictive maintenance applications in Industry 4.0 environments. We discuss mitigation strategies, edge/fog preprocessing, adaptive sampling, compression, and low-latency protocols, to preserve responsiveness in latency-sensitive deployments. The study offers a reproducible reference for aligning architectural choices with industrial requirements and provides concrete avenues to extend performance beyond 2,000 records s⁻¹ in future work.