This is an outdated version published on 2026-02-25. Read the most recent version.

AI-Driven Fairness-Aware Resource Scheduling and Optimization in Constrained Cyber-Physical Infrastructures

Authors

  • Wei Zhang Department of Electronic Engineering, Tsinghua University, China
  • Li Chen

DOI:

https://doi.org/10.66280/ijair.v1i1.9

Keywords:

cyber-physical systems; resource scheduling; fairness; multi-objective optimization; edge computing; reinforcement learning

Abstract

Resource scheduling in constrained cyber-physical infrastructures (e.g., industrial production systems, distribution IoT, and edge collaboration networks) must jointly satisfy latency, energy, reliability, and fairness objectives.  Static optimization is often insufficient under rapidly chang- ing workloads and system states.  This paper proposes an AI-driven fairness-aware scheduling framework that uses learning-assisted demand prediction and state estimation to guide multi- objective scheduling within feasibility constraints, while enforcing explicit fairness constraints to prevent long-term resource bias under high load.  We construct a unified task–resource model with timing and energy constraints, define fairness objectives based on the Jain index and min- imum satisfaction,  and introduce a hierarchical solution strategy  (prediction–planning–online correction).  Across simulated and scenario-based comparisons, the proposed method improves fairness and tail latency while meeting deadline and energy constraints, demonstrating scalabil- ity and practical applicability for multi-constraint CPS infrastructures.

Downloads

Published

2026-02-25

Versions

How to Cite

Zhang, W., & Chen, L. (2026). AI-Driven Fairness-Aware Resource Scheduling and Optimization in Constrained Cyber-Physical Infrastructures. International Journal of Artificial Intelligence Research, 1(1). https://doi.org/10.66280/ijair.v1i1.9