Smart Decision-Making and Communication Strategy in Industrial Internet of Things

Abstract

Smart machine-machine (M2M) interactions, such as those enabled by the Internet of Things (IoT), have enabled people and machines to communicate and make decisions together. Furthermore, these systems have become increasingly important in the commercial and industrial sectors over the previous two decades. The Industrial Internet of Things (IIoT) is a smart system comprising engineering equipment which can connect to one another to improve manufacturing operations. This task would become more complicated if the amount of energy used by the IIoT ecosystems, as well as the amount of network traffic they generate, increased dramatically. Consequently, decision-making processes during communication are essential for autonomous interaction in critical IoT infrastructure. Smart factories employ communication technology to track and gather information in real-time to enhance the output, effectiveness, and predictability while lowering the overall cost of vital operations. In this context, Industry 4.0 not only limits to addresses the issues of integrating technologies, but it also focuses on data collection, dissemination, utilization, and organization and also improves the delivery of the solution or services quicker with more sustainability. This study intends to create an NF-based communication system for IIoT platforms to leverage those benefits. The proposed model includes smart decision-making procedures to deal with communication issues. Compared with the many methods already in use, the suggested mechanism's functional viability in the automated system is found to be optimal. Outcomes from simulations reveal that the suggested method has improved the accuracy and communication reliability of the IIoT platforms in comparison with the previous methods. Aside from these, the suggested model keeps the throughput of the local automation unit at 96.03% and the throughput of the production hall at 95.58% on average while maintaining the lowest average PLR of about 26.48% across different data rates.

An Efficient Routing Algorithm for Implementing Internet-of-Things-Based Wireless Sensor Networks Using Dingo Optimizer

“…It is true that multiple real-world objects may be able to do the same task efficiently [17]. The availability of real devices, such as cars, buildings, and anything with actuators, sensors, implanted hardware, or programming, is made possible by the Internet of Things (IoT) [14]. These smart devices can collect and exchange data more skillfully for different purposes by enabling IoT [15].…”

“…The primary obstacle facing the Internet of Things is the limited availability of resources, such as electricity, processing power, memory, and wireless communication range and bandwidth, all of which have an impact on routing [13]. The availability of real devices, such as cars, buildings, and anything with actuators, sensors, implanted hardware, or programming, is made possible by the Internet of Things (IoT) [14]. These smart devices can collect and exchange data more skillfully for different purposes by enabling IoT [15].…”

CNAIS: Performance Analysis of the Clustering of Non-Associated Items Set Techniques

“…For huge datasets or a large number of clusters, K-Means can be computationally expensive [15][16][17][18][19][20]. Outliers can have a major impact on K-Means outcomes.…”

Optimizing Heterogeneity in IoT Infra Using Federated Learning and Blockchain-based Security Strategies

“…FedAvg eliminates inactive users based on their average response time, assuming that all users would participate at the same rate. Despite its widespread use, FedAvg still has a number of shortcomings when it comes to dealing with heterogeneity's challenges [12]. The researchers proposed a new approach called "SCAFFOLD"that gets usage management variables (diversity minimization) to reduce the negative impact of heterogeneity.…”