Optimal Resource Allocation Policy Research Articles

Dynamic resource allocation and scheduling for appointment-based systems with walk-ins

Effectively managing constrained and perishable capacity amid fluctuating demands is a common challenge in service industries, often addressed through appointment-based systems that incorporate walk-ins. Balancing appointments and walk-ins is crucial for enhancing system value and maintaining stability. In this article, we study an integrated decision problem for appointment scheduling and resource allocation across multiple facilities, considering stochastic demands from three segments: priority customers, walk-in customers, and appointment customers. We formulate the problem as a sequential scheduling-allocation challenge using stochastic dynamic programming. Leveraging the anti-multimodularity of the value function, we fully characterize the optimal dynamic resource allocation policy and its monotone structural properties under any appointment schedule. Given the complexity and intractability of optimal scheduling with dynamic resource allocation, we introduce two suboptimal scheduling policies—vertical scheduling and horizontal scheduling—and a heuristic allocation policy. Additionally, we create an Upper Bound (UB) for the original problem as a surrogate, establishing an asymptotically optimal UB for the scaled problem through its objective value and a corresponding lower bound through its solution. This UB solution serves as an efficient and effective heuristic, demonstrating superior performance compared with the combined performance of suboptimal scheduling policies with optimal dynamic resource allocation, especially as problem size increases. It also holds significant practical implications for integrated pooling systems.

A budgeting resource allocation model for capacity expansion

The reconfiguration of the global supply chain has created opportunities for newly industrialized economies (NIEs) to strengthen and expand their production capacities. We develop a dynamic modeling framework to study a firm’s resource allocation decisions while building capacity for a portfolio of production lines over multiple decision periods. Our model adds to the literature on versions of this problem in which distributions of market demands are incorporated into the investment–return function, and where no partial order fulfillment is possible. We consider two specific cases: one involving exogenously defined demand distributions, and one where the distributions of the parameters of the demand distribution are updated based on Bayesian inference. For each case, we first identify optimal capacity levels as a function of unit investment cost, holding cost, and the demand distribution of each production line. We then characterize the optimal path to reach these levels under different market conditions by deriving closed-form optimal resource allocation policies.

Uncovering Current and Future Variations of Irrigation Water Use Across China Using Machine Learning

AbstractAccurately characterizing changes in irrigation water use (IWU) is crucial for formulating optimal water resource allocation policies, particularly in the context of climate change. However, existing IWU estimation methods suffer from uncertainties due to limited data availability and model constraints, restricting their applicability on a national scale and under future climate change scenarios. We present a robust framework leveraging machine learning and multiple data sets to estimate IWU across China. Forced with an ensemble of climate and socio‐economic projections, we appraise future trends and additional costs of IWU. Our model shows high accuracy in reproducing IWU, with coefficient of determination (R2) ranging from 0.86 to 0.91 and root mean square error from 0.261 to 0.361 km3/yr when compared to reported values in Chinese prefectures. Independent validation at 11 cropland sites further confirms the model's predictive power (R2 = 0.67). Under different emissions scenarios, China's IWU is projected to increase by 8.5%–17.1% (6.8%–34.8%) by 2050s (2100s) compared to the historical period (1981–2010), with higher emissions leading to more significant increases. This rise in IWU by 2050s (2100s) comes with an estimated additional cost of US $1.65–3.91 ($2.28–6.5) billion/year, highlighting the urgency for sustainable water management. Our study provides an effective approach for estimating current and future IWU using machine learning techniques, transferable to other countries facing increasing irrigation demands.

Optimizing data processing for edge-enabled IoT devices using deep learning based heterogeneous data clustering approach

- Edge-based Internet of Things devices have transformed smart farming, aiding in efficient data collection and processing for optimal resource utilization and crop yields. However, task scheduling and resource allocation pose significant challenges due to the dynamic nature of agricultural environments. Our research introduces a novel framework that integrates deep reinforcement learning algorithm into an edge-enabled wireless sensor network for multi-objective optimization of the functionality of the Deep Q-Networks (DQNs). This framework extends the traditional Q-learning method to manage large state-action spaces efficiently. It employs a deep neural network to approximate the Q-value function, rather than relying on a Q-table, making it more capable of handling complex problems with high-dimensional state spaces. It forms heterogenous data clusters supports an optimal task scheduling and resource allocation policies, sustains key objectives such as minimal energy consumption, latency, efficient resource utilization, and reduced task completion time. The framework's performance is evaluated in a simulated environment mimicking real-world smart farming applications. Results confirm its superiority in enhancing performance metrics and lowering energy consumption, as opposed to traditional networks.

Blockchain-Empowered Resource Allocation in Multi-UAV-Enabled 5G-RAN: A Multi-Agent Deep Reinforcement Learning Approach

In 5G and B5G networks, real-time and secure resource allocation with the common telecom infrastructure is challenging. This problem may be more severe when mobile users are growing and connectivity is interrupted by natural disasters or other emergencies. To address the resource allocation problem, the network slicing technique has been applied to assign virtualized resources to multiple network slices, guaranteeing the 5G-RAN quality of service. Moreover, to tackle connectivity interruptions during emergencies, UAVs have been deployed as airborne base stations, providing various services to ground networks. However, this increases the complexity of resource allocation in the shared infrastructure of 5G-RAN. Therefore, this paper proposes a dynamic resource allocation framework that synergies blockchain and multi-agent deep reinforcement learning for multi-UAV-enabled 5G-RAN to allocate resources to smart mobile user equipment (SMUE) with optimal costs. The blockchain ensures the security of virtual resource transactions between SMUEs, infrastructure providers (InPs), and virtual network operators (VNOs). We formulate a virtualized resource allocation problem as a hierarchical Stackelberg game containing InPs, VNOs, and SMUEs, and then transform it into a stochastic game model. Then, we adopt a Multi-agent Deep Deterministic Policy Gradient (MADDPG) algorithm to solve the formulated problem and obtain the optimal resource allocation policies that maximize the utility function. The simulation results show that the MADDPG method outperforms the state-of-the-art methods in terms of utility optimization and quality of service satisfaction.

An intelligent approach of task offloading for dependent services in Mobile Edge Computing

With the growing popularity of Internet of Things (IoT), Mobile Edge Computing (MEC) has emerged for reducing the heavy workload at the multi-cloud core network by deploying computing and storage resources at the edge of network close to users. In IoT, services are data-intensive and event-driven, resulting in extensive dependencies among services. Traditional task offloading schemes face significant challenges in the IoT scenario with service dependencies. To this end, this paper proposes an intelligent approach for minimizing latency and energy consumption which jointly considers the task scheduling and resource allocation for dependent IoT services in MEC. Specifically, we establish the system model, communication model as well as computing model for performance evaluation by fully considering the dependent relationships among services, and an optimization problem is proposed for minimizing the delay and energy consumption simultaneously. Then, we design a layered scheme to deal with the service dependencies, and present detailed algorithms to intelligently obtain optimal task scheduling and resource allocation policies. Finally, simulation experiments are carried out to validate the effectiveness of the proposed scheme.

Joint Task Offloading and Resource Allocation for Device-Edge-Cloud Collaboration With Subtask Dependencies

With more computational intensive applications deployed involving mobile edge computing (MEC), the collaboration among mobile devices, edge and cloud servers becomes an effective mechanism to fully utilize all available distributed computing resources. However, two main challenges have yet to be addressed to enable this three-way collaboration for securing necessary computational resources and further guaranteeing the quality of service (QoS) of task handling. The first challenge is related to the partitioning of an application task into several dependent subtasks and schedule them among the collaborating device-edge-cloud (DEC). The second is focused on the allocation of necessary computing resources of device, edge and cloud servers for effective subtask handling. To this end, we study the joint task offloading and resource allocation for DEC collaboration in this paper by formulating a new optimization problem with the objective of minimizing the task handling latency. To solve this problem, we decompose the original problem into two subproblems, which include the first one of calculating the optimal task partitioning ratio by mathematical analytical method, as well as the second on using the Lagrangian dual (LD) method for obtaining the optimal task offloading and resource allocation policy. Finally, we conduct simulation experiments on a real-life dataset obtained from the central business district (CBD) of Melbourne, Australia, and the experimental results validate the efficacy of our approach in minimizing latency.

An Efficient Distributed Machine Learning Framework in Wireless D2D Networks: Convergence Analysis and System Implementation

Facing the heavy traffic burden and data privacy, distributed machine learning (DML) has been envisioned as a promising computing paradigm to enable edge intelligence by extracting and establishing models collaboratively in wireless networks. Unlike centralized methods, DML involves frequent local model exchanging among distributed devices, which confronts low efficiency referring to the convergence rate and delay. To enable edge intelligence, how to train DML efficiently has recently attracted extensive interest in wireless networks. Rather than over a fixed or centralized topology, we study the convergence and system implementation of DML over a wireless device-to-device (D2D) network in this paper. First, we introduce the DML training process and system model in this network, where the total delay in reaching convergence is minimized. Second, we analyze the convergence rate to figure out the special effects of synchronization frequency and network topology. To improve the efficiency of DML training, we propose a system implementation approach to reduce the convergence rate and delay of one iteration, where the network topology and synchronization frequency are set, followed by an optimal resource allocation policy. At last, we perform experiments with an image classification task. Simulation results indicate that our proposed D2D framework can effectively reduce training delay and promote computation efficiency by 39% in a heterogeneous environment, which is concordant with the theoretical analysis.

Joint Optimization for MEC Computation Offloading and Resource Allocation in IoV Based on Deep Reinforcement Learning

In the Internet of Vehicle (IoV), the limited computing capacity of vehicles hardly processes the intensive computation tasks locally. The computation tasks can be offloaded to multiaccess edge computing (MEC) servers for processing, where MEC provides the required computing capacity to the nearby vehicles. In this paper, we consider a scenario where there are cooperation and competition between vehicles, the offloading decision of any vehicle will affect the decisions of the others, and the computing resource allocation strategies by MEC will dynamically change. Therefore, we propose a joint optimization scheme for computation offloading decisions and computing resource allocation based on decentralized multiagent deep reinforcement learning. The proposed scheme learns the optimal actions to minimize the total weighted cost which is designed as the vehicles’ satisfaction based on the type of stochastic arrival tasks and dynamic interaction between MEC server and vehicles within different RSUs coverages. The numerical results show that the proposed algorithms based on decentralized multiagent deep deterministic policy gradient (DDPG) which is named De-DDPG can autonomously learn the optimal computation offloading and resource allocation policy without a priori knowledge and outperform the other three baseline algorithms in terms of the rewards.

Resource Allocation of Hybrid VLC/RF Systems With Light Energy Harvesting

In this paper, we study the problem of maximizing the sum throughput of <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">${K}$ </tex-math></inline-formula> users in a hybrid heterogeneous visible light communication (VLC)/radio frequency (RF) wireless communication system. This is done by optimizing the transmission time intervals of a time division multiple access (TDMA) scheme allocated to the users in two different downlink (DL) scenarios. In both scenarios, using the harvested energy, each user transmits its signal in an optimal allocated time interval over the uplink (UL) channel. In the first scenario, a lightwave powered communication network (LPCN)-based VLC system is used to radiate only optical energy to be harvested by the users over DL channel. Specifically, UL sum-rate maximization problem is formulated by optimizing UL time allocations. In the second scenario, a time switching-based simultaneous lightwave information and power transfer (TS-based SLIPT) is considered where the light-emitting diode (LED) transmitter sends both information and power simultaneously over DL channel. Specifically, we propose a multi-objective optimization problem (MOOP) to study the trade-off between the UL and DL sum-rates. The non-convex MOOP framework is then transformed into an equivalent form, which yields a set of Pareto optimal resource allocation policies. The effectiveness of the proposed approaches is illustrated through numerical results.

Multi-Agent DRL for Task Offloading and Resource Allocation in Multi-UAV Enabled IoT Edge Network

The Internet of Things (IoT) edge network has connected lots of heterogeneous smart devices, thanks to unmanned aerial vehicles (UAVs) and their groundbreaking emerging applications. Limited computational capacity and energy availability have been major factors hindering the performance of edge user equipment (UE) and IoT devices in IoT edge networks. Besides, the edge base station (BS) with the computation server is allowed massive traffic and is vulnerable to disasters. The UAV is a promising technology that provides aerial base stations (ABSs) to assist the edge network in enhancing the ground network performance, extending network coverage, and offloading computationally intensive tasks from UEs or IoT devices. In this paper, we deploy a clustered multi-UAV to provide computing task offloading and resource allocation services to IoT devices. We propose a multi-agent deep reinforcement learning (MADRL)-based approach to minimize the overall network computation cost while ensuring the quality of service (QoS) requirements of IoT devices or UEs in the IoT network. We formulate our problem as a natural extension of the Markov decision process (MDP) concerning stochastic game, to minimize the long-term computation cost in terms of energy and delay. We consider the stochastic time-varying UAVs’ channel strength and dynamic resource requests to obtain optimal resource allocation policies and computation offloading in aerial to ground (A2G) network infrastructure. Simulation results show that our proposed MADRL method reduces the average costs by 38.643%, and 55.621% and increases the reward by 58.289% and 85.289% compared with the different single agent DRL and heuristic schemes, respectively.

Delay-Sensitive Joint Optimal Control and Resource Management in Multiloop Networked Control Systems

In the operation of networked control systems (NCSs), where multiple processes share a resource-limited and time-varying cost-sensitive network, communication delay is inevitable and primarily induced by, first, intermittent sensor sampling to restrict nonurgent transmissions, and second, resource management to avoid contentions, excessive traffic, and data loss. In a heterogeneous scenario, where control systems may tolerate only specific levels of sensor-to-controller latency, delay sensitivities need to be considered in the design of control and network policies to achieve the desired performance guarantees. We propose a cross-layer optimal co-design of control, sampling, and resource management policies for an NCS consisting of multiple stochastic linear time-invariant systems which close their sensor-to-controller links over a shared network. Aligned with advanced communication technology, we assume that the network offers a range of latency-varying transmission services for given prices. The performance of the local closed-loop systems is measured by a combination of linear-quadratic Gaussian cost and a suitable communication cost, and the overall objective is to minimize a defined social cost by all three policymakers. We derive optimal control, sampling, and resource allocation policies under different cross-layer awareness models, including constant and time-varying parameters, and show that higher awareness generally leads to performance enhancement at the expense of higher computational complexity. This trade-off is shown to be a key feature to select the proper interaction structure for the codesign.

On the Application of BAC-NOMA to 6G umMTC

This letter studies the application of backscatter communications (BackCom) assisted non-orthogonal multiple access (BAC-NOMA) to the envisioned sixth-generation (6G) ultra-massive machine type communications (umMTC). In particular, the proposed BAC-NOMA transmission scheme can realize simultaneous energy and spectrum cooperation between uplink and downlink users, which is important to support massive connectivity and stringent energy constraints in umMTC. Furthermore, a resource allocation problem for maximizing the uplink throughput and suppressing the interference between downlink and uplink transmission is formulated as an optimization problem and the corresponding optimal resource allocation policy is obtained. Computer simulations are provided to demonstrate the superior performance of BAC-NOMA.

Optimal Resource Allocations for Statistical QoS Provisioning to Support mURLLC Over FBC-EH-Based 6G THz Wireless Nano-Networks

One of most important techniques for enabling the sixth-generation (6G) mobile wireless network lies in how to efficiently guarantee various stringent quality-of-service (QoS) performance-metrics to support the emerging massive Ultra-Reliable Low-Latency Communications (mURLLC) in 6G. Correspondingly, finite blocklength coding (FBC) has been developed as an effective technique to significantly improve various QoS indices for mURLLC through implementing short-packet communications. On the other hand, Terahertz (THz) band wireless nano-communications have been widely envisioned as a promising 6G technique to efficiently support utra-high data-rate (up to 1 Tbps). One of the major constraints over THz-band nano-networks is the severely limited energy that can be accessed by nano devices. Towards this end, various novel energy harvesting (EH) mechanisms have been proposed to remedy the energy scarcity problem. However, how to accurately characterize the relationships among THz wireless channels, energy consumption, and EH models for FBC based nano communications remains a challenging problem to support statistical delay and error-rate bounded QoS provisioning over FBC based 6G THz wireless nano-networks. To overcome these challenges, in this paper we propose optimal resource allocation policies to achieve the maximum ε-effective capacity in the THz band over FBC-EH-based nano-networks. Particularly, we establish nano-scale system models and characterize wireless channel models in the THz band using FBC. In order to support statistical delay and error-rate bounded QoS provisioning, we formulate and solve the ε-effective capacity maximization problem under several different EH constraints for our proposed schemes. Simulation results are included, which validate and evaluate our proposed schemes in the finite blocklength regime.

Resource Allocation for Open-Loop Ultra-Reliable and Low-Latency Uplink Communications in Vehicular Networks

To support ultra-reliable and low-latency communication (URLLC) in vehicular networks, the virtual cells, where multiple access points (APs) cooperatively serve one mobile node, have been proposed to reduce the end-to-end latency in the downlink. The latency can be further reduced by eliminating the need for retransmission and feedback control, i.e., open-loop communications. However, it is difficult to achieve a high reliability of the uplink via virtual cells, because of multiple access interference and collisions from other virtual cells nearby. In this paper, we formulate the proactive radio resource allocation in the open-loop uplink of vehicular networks as a stochastic optimization problem with the objective to maximize the uplink reliability while ensuring the network stability. The optimal resource allocation policies are obtained solving the optimization problem using the Lyapunov optimization technique in a distributed manner. To reduce the computational and to improve the challenges of the Lyapunov optimization, we propose a virtual resource slicing algorithm that maps radio resource units to virtual resource blocks. Simulation results exhibit that both ultra-low latency and ultra high reliability are guaranteed in the open-loop uplink of vehicular networks. Based on the theoretical performance analysis and simulations, the uplink radio access procedure is summarized for the URLLC in vehicular networks.

Fair Scheduling of Heterogeneous Customer Populations

When managing congested service systems, it is common to use priority rules based on some operational criteria. In this paper, we consider the societal implications of such individual-focused priority policies, when individuals are considered as members of broader population groups. We find that optimal resource allocation policies such as the cμ-rule in scheduling can lead to significant inequity across different population groups. We propose policies that can mitigate this inequity and can even generate completely equitable outcomes across populations with little, or at times, even no additional system cost. Thus, we find that it can be possible to achieve more equitable outcomes while ensuring operational efficiency.

Performance Analysis on Cache-Enabled FR2 IAB Networks

This paper brings the concept of wireless edge caching (WEC) to the realm of integrated access and backhaul (IAB) networks. Traditionally, both access and backhaul are allocated fixed spectrum. However, when WEC is applied to the IAB nodes to pre-fetch some popular files, the backhaul traffic can be drastically reduced. Accordingly, an analytical framework is developed for a cache-enabled IAB network for the wide-band frequency range 2 (FR2) (> 24.25 GHz) channel model. In particular, the optimal spectrum resource allocation policy between access and backhaul for IAB networks is investigated by utilizing WEC. For the considered WEC enabled IAB network, a mathematical framework is presented under the content delivery phase for the calculation of three key performance metrics, namely the average success probability (ASP), throughput, and latency of file delivery with respect to various network parameters such as different caching placement strategies, node density, cache size, antenna number, blockage density, and resource partitioning factor. Extensive numerical results are provided to both verify the theoretical derivations and to demonstrate the precedency of using WEC as a cost-effective measure for the improvement of the IAB network’s performance. In particular, under a normalized cache size of 0.6, compared to the baseline no caching scenario, the percentages of backhaul spectrum that can be saved are 48% (26%), 66% (48%), and 47% (27%) when caching the most popular files (caching uniformly) with respect to the ASP, latency, and throughput of file delivery, respectively. The saved backhaul spectrum can then be shifted to the access for its performance enhancements. Our findings also show that there exists an optimal spectrum partition for IAB with respect to different network parameter settings. Further, there exists a tradeoff in the selection of the optimal partitioning factor with respect to ASP/ throughput and latency of file delivery for varying IAB node densities.

Deep Learning for Radio Resource Allocation With Diverse Quality-of-Service Requirements in 5G

To accommodate diverse Quality-of-Service (QoS) requirements in 5th generation cellular networks, base stations need real-time optimization of radio resources in time-varying network conditions. This brings high computing overheads and long processing delays. In this work, we develop a deep learning framework to approximate the optimal resource allocation policy that minimizes the total power consumption of a base station by optimizing bandwidth and transmit power allocation. We find that a fully-connected neural network (NN) cannot fully guarantee the QoS requirements due to the approximation errors and quantization errors of the numbers of subcarriers. To tackle this problem, we propose a cascaded structure of NNs, where the first NN approximates the optimal bandwidth allocation, and the second NN outputs the transmit power required to satisfy the QoS requirement with given bandwidth allocation. Considering that the distribution of wireless channels and the types of services in the wireless networks are non-stationary, we apply deep transfer learning to update NNs in non-stationary wireless networks. Simulation results validate that the cascaded NNs outperform the fully connected NN in terms of QoS guarantee. In addition, deep transfer learning can reduce the number of training samples required to train the NNs remarkably.

Parametric transformation of timed weighted marked graphs: applications in optimal resource allocation

Timed weighted marked graphs are a subclass of timed Petri nets that have wide applications in the control and performance analysis of flexible manufacturing systems. Due to the existence of multiplicities (i.e., weights) on edges, the performance analysis and resource optimization of such graphs represent a challenging problem. In this paper, we develop an approach to transform a timed weighted marked graph whose initial marking is not given, into an equivalent parametric timed marked graph where the edges have unitary weights. In order to explore an optimal resource allocation policy for a system, an analytical method is developed for the resource optimization of timed weighted marked graphs by studying an equivalent net. Finally, we apply the proposed method to a flexible manufacturing system and compare the results with a previous heuristic approach. Simulation analysis shows that the developed approach is superior to the heuristic approach.

Optimal Resource Allocation for MC-NOMA in SWIPT-Enabled Networks

In this letter, we study a receiver architecture technique for joint resource allocation in a downlink (DL) multi-user multi-carrier non-orthogonal multiple access (MC-NOMA) network with simultaneous wireless information and power transfer (SWIPT). In this framework, the subcarrier set is partitioned into two groups that are assigned to perform information decoding (ID) and energy harvesting (EH) at the receiver side based on the optimization problem. This letter seeks to maximize energy harvesting while meeting a minimum data-rate requirement for each user. The underlying optimization problem is mixed-integer non-linear programming (MINLP). To that end, we employ the monotonic optimization approach to obtain an optimal resource allocation policy. A suboptimal solution is also presented. Simulation results demonstrate that our proposed algorithm achieves excellent performance as compared to other works described in the literature.