A newly proposed multi-objective genetic algorithm improves the task completion rate for delay-sensitive Internet of Things tasks by 28.6 percent compared with the classic Non-dominated Sorting Genetic Algorithm II, according to experimental results published in heterogeneous edge computing research. The model, termed the multi-objective genetic algorithm based on hierarchical probability selection and priority scheduling, or HP-MOGA, also reduced average task cost by 7.3 percent in the same experimental evaluations. By simultaneously optimizing multiple performance parameters, the computational framework provides an efficient operational solution for scheduling diverse workloads across distributed computing infrastructures.
The investigation addresses severe operational challenges that arise when scheduling tasks across end-edge-cloud heterogeneous environments. In such systems, Internet of Things tasks are integrated directly with edge computing technology to process demanding computational workloads. Scheduling in this context is complicated by the marked heterogeneity of incoming tasks and available computing resources, alongside strict real-time requirements.
To resolve these scheduling conflicts, the study formulates task allocation as a multi-objective optimization problem. The formulation explicitly identifies three simultaneous optimization objectives: improving the overall task completion rate, lowering the average task cost, and reducing the average task delay.
To implement this multi-objective framework, HP-MOGA introduces two core algorithmic mechanisms: a priority scheduling mechanism based on laxity time and a hierarchical probability selection strategy.
Subject of Research: Task scheduling optimization for delay-sensitive Internet of Things applications in heterogeneous edge computing
Article Title: Cost-efficient scheduling for delay-sensitive IoT tasks in heterogeneous edge computing
Article References: Liu, X., Liu, Z., Hou, S., Zhao, Y., & Tong, Q. (2027). Cost-efficient scheduling for delay-sensitive IoT tasks in heterogeneous edge computing. Cluster Computing, 30(1), Article 50. https://doi.org/10.1007/s10586-026-06754-x
Image Credits: AI Generated
DOI: 10.1007/s10586-026-06754-x
Keywords: heterogeneous edge computing, task scheduling, delay-sensitive tasks, Internet of Things, HP-MOGA, multi-objective genetic algorithm, hierarchical probability selection, priority scheduling, laxity time, NSGA-II, Pareto optimal solutions, task completion rate, average task cost, average task delay, end-edge-cloud environment, multi-objective optimization, resource heterogeneity
Cite Scienmag News
Marilyn Langley. (November 19, 2026). Priority Scheduling Algorithm Improves Edge Computing Task Completion by 28 Percent. Scienmag. https://scienmag.com/priority-scheduling-algorithm-improves-edge-computing-task-completion-by-28-percent/
Marilyn Langley. "Priority Scheduling Algorithm Improves Edge Computing Task Completion by 28 Percent." Scienmag, 19 November 2026, https://scienmag.com/priority-scheduling-algorithm-improves-edge-computing-task-completion-by-28-percent/. Accessed 19 November 2026.
Marilyn Langley. "Priority Scheduling Algorithm Improves Edge Computing Task Completion by 28 Percent." Scienmag. November 19, 2026. https://scienmag.com/priority-scheduling-algorithm-improves-edge-computing-task-completion-by-28-percent/








