Rendezvous optimization method for addressing long-job burst problem in industrial internet of things
Year: 2026
Type: article
Abstract: Industrial Internet of Things (IIoT) cells suffer long-job bursts when arrival surges and interdependent tasks push jobs past their time limit. This article presents ROM-RS, a Rendezvous Optimization Method with rule-based Recommendation Scheduling for adaptive task scheduling and resource reallocation. ROM-RS matches a resource to a waiting task at the moment the resource is released, reallocates released resources across jobs, and applies three runtime rules that favour the jobs that can still finish within the limit. The method is evaluated on a discrete-event simulation of a CNC cell whose task types, precedence and processing times are taken from a public machining dataset, with 1,080 jobs per 8-hour shift and arrival surges, against first-come first-served dispatching, PSO- and GA-tuned priority rules and a DQN rule-selection scheduler. ROM-RS yields the lowest burst ratio of the five methods (0.125 against 0.150 for the tuned rules and 0.301 for FCFS; paired p < 0.001) with 0.25 s of computation per shift, while the tuned rules complete about 0.5% points more jobs at roughly 600 times the computation. An ablation attributes the result to the recommendation rules, the event-driven matching and the reallocation of released resources. The simulator and all outputs are deposited.
Source: Scientific Reports
Institutions: University of Tabuk, University of Bisha
Language: English
Subfield: Industrial and Manufacturing Engineering
Field: Engineering
Domain: Physical Sciences
Open Access status: gold
Estimated APC paid: $2,690
Funder: University of Tabuk
OpenAlex ID: W7228535212
