Efficient optimization techniques for resource allocation in UAVs mission framework

Razzaq, Sohail; Xydeas, C.S.; Mahmood, Anzar; Ahmed, Saeed; Ratyal, Naeem Iqbal; Iqbal, Jamshed · 2023 · OpenAlex-citations

DOI: 10.1371/journal.pone.0283923

archive: archived pipeline: cataloged verified

Get this paper ↗ (DOI — opens at the source; we link to it, we don't host it)

Summary

This paper addresses the problem of efficient resource allocation for Unmanned Aerial Vehicles (UAVs) within a mission framework, specifically modeled as a firefighting operation. The core research question involves selecting an optimal subset of UAVs from a larger pool of candidates to maximize mission performance while adhering to resource constraints. The authors frame this as a discrete, nonlinear optimization problem with binary variables, where each variable represents the inclusion or exclusion of a specific UAV. The study is motivated by the need for computationally efficient solutions for time-critical applications, extending beyond firefighting to other domains such as healthcare, surveillance, and wireless communications. The methodology develops a comprehensive UAV firefighting mission framework that includes a Mission Model (MM) and a Mission Performance Objective Measure (MPOM). The MM models fire intensity as a function of time, which decreases based on the amount and effectiveness of delivered suppression resources. The MPOM is defined as the area under the fire intensity curve, representing the extent of fire damage; minimizing this value constitutes the optimization objective. The authors propose and compare both deterministic and stochastic optimization techniques. Deterministic methods assume fixed input variables, while stochastic methods account for uncertainty in resource delivery effectiveness by modeling variables as Gaussian random variables. To handle stochastic constraints without excessive computational cost, the authors introduce penalty terms into the objective function proportional to the expected magnitude of constraint violations. The study evaluates both Exhaustive Optimization (EO), which guarantees global optimality but is NP-hard, and novel heuristic search schemes designed for computational efficiency. Experimental results derived from computer simulations demonstrate the performance of these techniques. The simulation environment accounts for UAV locations, flight paths, fuel consumption, and resource capacities across multiple bases. The findings indicate that the proposed stochastic multistage optimization schemes offer acceptable accuracy and significant computational efficiency compared to deterministic counterparts. The heuristic methods successfully identify suboptimal but effective solutions rapidly, making them suitable for real-time, time-critical scenarios where exhaustive search is intractable due to the exponential growth of the search space with the number of UAVs. The stochastic approach effectively manages uncertainty by penalizing solutions that likely violate resource constraints, thereby enhancing robustness. The significance of this work lies in the development of a generic resource allocation framework applicable to various UAV missions and other resource-constrained systems. By providing both deterministic and stochastic optimization techniques, the paper offers a balanced approach to trade-offs between solution quality and computational complexity. The introduction of penalty-based stochastic optimization allows for robust decision-making under uncertainty without the prohibitive costs of exact probabilistic calculations. These contributions advance the field by providing scalable, efficient algorithms for optimizing UAV deployments in dynamic and uncertain environments, supporting broader applications in emergency response, logistics, and network resource management.

Provenance

The full processing record for this entry. Every stage of this paper's journey through the pipeline is logged — what ran, with which tool and model, how many attempts it took, and when it last completed.

Summary generated by qwen3.6-27b-prismaquant on 2026-06-26; verification: verified.

Topics

Ranked by relevance to this paper. Hover a topic for its definition.

• driverless ads