Artificial Intelligence-Based Predictive Network Congestion Control for Mission-Critical Industrial Communication Systems

One of the biggest problems currently hindering reliable and consistent performance for any type of industrial communication system, such as time-sensitive control subsystems seen in the oil & gas industry, is network congestion. This study aims to create an artificial intelligence (AI) predictive model to detect congestion events in the Oil Company (OC) network. To investigate the performance of the Random Forest, XGBoost, and LSTM models for network congestion detection, this study uses a quantitative comparative experimental design involving 1,000 operational network records from a network of an oil company. They experimented with three models based on machine learning as shown below: Random Forest, XGBoost and LSTM. The results revealed that all the models were performing with a very high level of accuracy ( Accuracy>96%). Indeed, XGBoost performed best among all the models evaluated in terms of all the measures we used, with Recall = 0.978 and F1-Score = 0.966, a good match for mission-critical environments that may encounter misdiagnosed traffic congestion and disrupt industrial processes. The results indicate that when AI is used to predict congestion in high-quality digital infrastructures, it can be beneficial in several areas, such as QoS (Quality of Service) improvement, packet loss mitigation, and predictive maintenance. The main value of this study is the suggested actionable framework specifically for oil-sector networks and highlighting new trends integrating AI with Digital Twin & Edge Technologies. The results could be used to guide the choice of congestion-detection models to track the performance of communication in SCADA-enabled industrial networks

Authors

Institutions

Publication Details

Journal
Scientific journal of engineering and technology.
Published
2026-11-14
DOI
https://doi.org/10.69739/sjet.v3i2.2520
Primary Topic
Network Traffic and Congestion Control
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

Artificial Intelligence-Based Predictive Network Congestion Control for Mission-Critical Industrial Communication Systems

Taha Abdulwahid Mahmood, Layla Mahdi Salih, Afnan waleed Abdulsattar
Scientific journal of engineering and technology.
Network Traffic and Congestion Control
article

Artificial Intelligence-Based Predictive Network Congestion Control for Mission-Critical Industrial Communication Systems

Taha Abdulwahid Mahmood, Layla Mahdi Salih, Afnan waleed Abdulsattar
article en

Abstract

One of the biggest problems currently hindering reliable and consistent performance for any type of industrial communication system, such as time-sensitive control subsystems seen in the oil & gas industry, is network congestion. This study aims to create an artificial intelligence (AI) predictive model to detect congestion events in the Oil Company (OC) network. To investigate the performance of the Random Forest, XGBoost, and LSTM models for network congestion detection, this study uses a quantitative comparative experimental design involving 1,000 operational network records from a network of an oil company. They experimented with three models based on machine learning as shown below: Random Forest, XGBoost and LSTM. The results revealed that all the models were performing with a very high level of accuracy ( Accuracy>96%). Indeed, XGBoost performed best among all the models evaluated in terms of all the measures we used, with Recall = 0.978 and F1-Score = 0.966, a good match for mission-critical environments that may encounter misdiagnosed traffic congestion and disrupt industrial processes. The results indicate that when AI is used to predict congestion in high-quality digital infrastructures, it can be beneficial in several areas, such as QoS (Quality of Service) improvement, packet loss mitigation, and predictive maintenance. The main value of this study is the suggested actionable framework specifically for oil-sector networks and highlighting new trends integrating AI with Digital Twin & Edge Technologies. The results could be used to guide the choice of congestion-detection models to track the performance of communication in SCADA-enabled industrial networks

Scientific journal of engineering and technology.Vol. 3(2)
University of Kirkuk (IQ)
Network Traffic and Congestion Control
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon