Adaptive mutation optimization algorithm for applications pre-deployment in IIoT edge devices

doi:10.1016/j.jfranklin.2026.108821

Adaptive mutation optimization algorithm for applications pre-deployment in IIoT edge devices

2026-06-16

0

PRE

AI

DOI：10.1016/j.jfranklin.2026.108821

Original

Original request for help

Save

Abstract

En 中文

The Industrial Internet of Things (IIoT) connects heterogeneous devices, forming a large-scale complex network. Deploying diverse applications onto resource-constrained edge devices creates a combinatorial optimization problem, influenced by memory limits, application dependencies, and dynamic application values. To solve this, we propose an intelligent pre-deployment mechanism using an adaptive mutation cuckoo search (A-MCS) algorithm. This approach automates the selection of an optimal application portfolio for edge devices. The A-MCS method improves metaheuristic optimization by combining mutation operators with an adaptive parameter control framework for dynamic self-tuning. We establish a mathematical model for the application selection problem and implement a prototype system to simulate IIoT edge environments. Experimental results show that A-MCS outperforms standard heuristic algorithms under various memory constraints and uncertain data conditions. These results underscore A-MCS’s potential to enhance resource allocation and operational intelligence in IIoT systems.

Keywords:

Metaheuristic algorithm

Cuckoo search

Parameter adaptation

Industrial internet of things

Resource allocation

Adaptive mutation optimization algorithm for applications pre-deployment in IIoT edge devices

Abstract

Journal

Organization