Multi-Objective Task-Aware Offloading and Scheduling Framework for Internet of Things Logistics

Area of Science:

• * Internet of Things (IoT)
• * Smart Transportation Systems
• * Cloud and Fog Computing

Background:

• * IoT devices in smart transportation have limited computational power, necessitating task offloading to remote servers.
• * Existing task offloading mechanisms often overlook fault tolerance, data awareness, and multi-parameter node selection for logistics.
• * This gap hinders optimal performance in energy consumption and response time for IoT logistics trucks.

Purpose of the Study:

• * To propose a novel Multi-Objective Task-Aware Offloading and Scheduling Framework for IoT Logistics (MT-OSF).
• * To address limitations in fault-tolerance, task/data awareness, priority-based offloading, and multi-parameter fog node selection.
• * To enhance the efficiency and reliability of IoT-based smart transportation.

Main Methods:

• * Developed a priority-based offloader to categorize tasks into delay-sensitive and computation-intensive.
• * Implemented a Task-Aware Scheduler (TAS) using the Analytical Hierarchy Process (AHP) for multi-criterion fog node selection (energy, bandwidth, RAM, MIPS).
• * Integrated a fault-tolerant manager for task re-execution and node failure handling, utilizing the First-Come, First-Served (FCFS) algorithm for scheduling.

Main Results:

• * Achieved a 7% reduction in response time compared to existing methods.
• * Demonstrated a 16% decrease in energy consumption.
• * Showcased a significant 22% reduction in the task failure ratio.

Conclusions:

• * The MT-OSF framework effectively optimizes task offloading and scheduling in IoT logistics.
• * The proposed system significantly improves performance metrics, including response time, energy efficiency, and fault tolerance.
• * MT-OSF offers a robust solution for enhancing the reliability and efficiency of smart transportation systems.