An Overview of Computational Models for Industrial Internet of Things to Enhance Usability

Abstract

In the last decade, the Internet of Things (IoT) has grown to connect a large number of smart entities, devices, and components. These connected entities provide a wide range of services to improve the current society of end customers. The Industrial Internet of Things (IIoTs) are revolutionary systems that have linked manufacturing processes with Internet access in order to preciously increase quality of services. These systems have minimized the costs of production through collaboration with electronic objects, accumulating computing, advanced analytics, and smart perception techniques. A demanding analysis of the strengths and limitations of computational models of IIoT is an essential part of the industry and before deciding which approach to use and implement for enhancing usability. Therefore, the goal of this study is to provide feedback and information to the research community and identify patterns in recommendations for future research in the context of process, development, and monitoring of additional technologies of computational models for IIoT. This paper has presented a comprehensive summary of the existing literature on IIoT for providing details about modern industrial revolutions in the context of IIoT. Associated materials were searched and filtered for identification of relevant materials to the proposed study. These materials have been collectively studied with in-depth analysis and then summarized to condense the information of computation models for the readers as well as entrepreneurs. The study will facilitate research community and practitioners to develop novel techniques, algorithms, and tools to automate and facilitate IIoT. This will develop the field of IIoT and will enhance its usability.

An Innovative Sensing Machine Learning Technique to Detect Credit Card Frauds in Wireless Communications

“…In recent years, the usage of DNN for credit card fraud detection has increased [16]. A few of them are autoencoder-based detection [17], K-means deep network [18], and LSTM using attention mechanism dependence concerns are better addressed by RNN architecture. Ability to learn order dependency in sequence prediction problems as a behaviour required in complicated issue areas like as machine translation and speech recognition, among others [19].…”

Optimization of Edge Resources for Deep Learning Application with Batch and Model Management

“…The edge computing architecture is now considered one of the major pillars of Industry 4.0, and is widely adopted in various applications from cloud-based IoT to the AI service [3]. Various approaches to realize cloud computing for Industry 4.0 are presented [4]. Object detection is one of the famous use cases in realization of Industry 4.0 with edge computing including defect detection, vehicle detection, and identification.…”

A Sensor Fusion Method Using Transfer Learning Models for Equipment Condition Monitoring

“…However, the new trend is to make the most out of sensor data and integrate it with traditional manufacturing execution systems. In this case, computational models for IoT system-generated data can add a significant value [1]. This might find important applications, especially in fault detection and equipment maintenance.…”