MAFC: Multi-Agent Fog Computing Model for Healthcare Critical Tasks Management

Area of Science:

• Healthcare technology
• Edge computing
• Artificial intelligence

Background:

• Healthcare generates massive data from sensors and devices, requiring efficient management.
• Fog computing offers edge network solutions but faces resource limitations in Fog Nodes.
• Optimizing task allocation in resource-constrained Fog Nodes is crucial for timely healthcare analytics.

Purpose of the Study:

• To propose a Multi-Agent Fog Computing (MAFC) model for effective management of critical healthcare tasks.
• To optimize task scheduling by considering task priority, network load, and Fog Node resource availability.
• To enhance decision-making for task allocation at the network edge.

Main Methods:

• Developed a Multi-Agent System (MAS) to manage critical healthcare tasks.
• Implemented a decision-making process involving three tables for task prioritization and allocation.
• Utilized the iFogSim simulator and UTeM clinic data for performance evaluation.

Main Results:

• The proposed MAFC model effectively maps tasks to Fog Nodes based on priority and resource availability.
• The system intelligently allocates tasks locally, to neighbor Fog Nodes, or to the Cloud.
• Demonstrated the applicability and optimality of the scheme in edge healthcare scenarios.

Conclusions:

• The MAFC model provides an optimized approach for critical task management in edge healthcare environments.
• Efficient resource utilization and task scheduling are achieved through the multi-agent decision-making process.
• The proposed solution addresses Fog Node resource limitations for improved healthcare data analytics.