Table 5 Summary of delay-minimizing scheduling schemes in adaptive and dynamic scheduling.

MADT-IoT⁵²

Minimizes long-term AoI using deep reinforcement learning and Lagrangian optimization

Low AoI, satisfies delay constraint, adaptive to transmission timeliness

Complex model, requires extensive computation

WISE⁵³

DRL-based wireless TSN scheduler that learns time-sensitive patterns for low-latency communication in IEEE 802.11 networks.

Meets up to 99.9% requirements, low delay under dynamic conditions, scalable

Needs exclusive channels, may be less energy-efficient

DSPA⁵⁴

Uses heuristic algorithms to schedule DSPA tasks in EFC environments, minimizing cost and delay

Cost-effective, adapts to changing capacity and deployment state

May struggle with extreme resource fluctuations

SPEA-II⁵⁵

Multi-objective optimization for minimizing delay and energy in task scheduling using evolutionary algorithms

Balances energy and response time, better resource use

Complex optimization may not scale well in real-time

IPAQ⁵⁶

Time-aware, multi-objective scheduling with particle swarm optimization and AHP to prioritize tasks

Handles large task volumes, prioritizes high-sensitivity data

Computationally intensive, depends on accurate classification

E2rC⁵⁷

migration technique for configuration of computational IoT.

deep reinforcement with a Markov multi-phase decisions with lower end-to-end delays and uses less energy

requires resources and complexity, degrade system performance

ISL-ITMH⁵⁸

energy optimization and trust management,

packet delivery ratio, routing overhead, and end-to-end delay

degrades system performance by multiple devices