← Back to Publications List

A Machine Learning-Based Edge–Cloud Framework for Intelligent System Optimization

Students & Supervisors

Student Authors
Shahriar Sadib
Bachelor of Science in Computer Science & Engineering, FST
Tanvir Hasan Samad
Bachelor of Science in Computer Science & Engineering, FST
Arizit Chaki Artha
Bachelor of Science in Computer Science & Engineering, FST
Supervisors
Md. Mortuza Ahmmed
Associate Professor, Faculty, FST

Abstract

Industrial Internet of Things (IIoT) is increasingly applied to smart manufacturing scenarios, where timely monitoring and intelligent decision-making are essential for achieving high levels of efficiency and integrity of the system. However, the increasing amount of sensory data and system complexity have made it difficult to detect faults and optimize performance. This study develops a machine learning approach for optimizing an intelligent system by analyzing real-world IIoT data in an edge-cloud setting. The data set contains various features from sensors as well as from systems, including temperature, pressure, vibration, latency in networks, and latency in edge computing. Exploratory data analysis is carried out in order to examine the features and their relationship to one another. In exploratory data analysis, the relationships among system latency, vibration, and the failure of circumstances were found to be highly correlated. A random forest classifier was used to build a predictive model, and the data set was split into a training and test set for evaluation purposes. Experimental analysis reveals that the accuracy of the model is 93.5% with an excellent precision value (0.99) in identifying failures and excellent recall (0.99) for normal system behavior. Moreover, in the aspect of feature importance, the results show that network latency, vibration, and edge processing times are the top influential features in system performance prediction. In all cases, the key objective is to evaluate the efficiency of machine learning when combined with the edge-cloud computing architecture. The edge devices provide timely predictions while the cloud-based systems perform deep data analysis. Therefore, it can be stated that the framework has high suitable potential.

Keywords

Industrial Internet of Things (IIoT) Machine Learning Edge-Cloud Computing

Publication Details

  • Type of Publication:
  • Conference Name: International Conference on Emerging Frontiers in Advanced Sciences and Technologies 2026
  • Date of Conference: 27/06/2026 - 27/06/2026
  • Venue: Pabna University of Science and Technology (PUST)
  • Organizer: Pabna University of Science and Technology (PUST) and Universiti Malaysia Perlis (UniMAP)
View on Publisher