Policy Deployment Research Articles

Energy-Efficient Edge Caching and Task Deployment Algorithm Enabled by Deep Q-Learning for MEC

Container technology enables rapid deployment of computing services, while edge computing reduces the latency of task computing and improves performance. However, there are limits to the types, number and performance of containers that can be supported by different edge servers, and a sensible task deployment strategy and rapid response to the policy is a must. Therefore, by jointly optimizing the strategies of task deployment, offloading decisions, edge cache and resource allocation, this paper aims to minimize the overall energy consumption of a mobile edge computing (MEC) system composed of multiple mobile devices (MD) and multiple edge servers integrated with different containers. The problem is formalized as a combinatorial optimization problem containing multiple discrete variables when constraints of container type, transmission power, latency, task offloading and deployment strategies are satisfied. To solve the NP-hard problem and achieve fast response for sub-optimal policy, this paper proposes an energy-efficient edge caching and task deployment policy based on Deep Q-Learning (DQCD). Firstly, the pruning and optimization of the exponential action space consisting of offloading decisions, task deployment and caching policy is completed to accelerate the training of the model. Then, the iterative optimization of the training model is completed using a deep neural network. Finally, the sub-optimal task deployment, offloading and caching policies are obtained based on the training model. Simulation results demonstrate that the proposed algorithm is able to converge the model in very few iterations and results in a great improvement in terms of reducing system energy consumption and policy response delay compared to other algorithms.

A Bi-Objective Learn-and-Deploy Scheduling Method for Bursty and Stochastic Requests on Heterogeneous Cloud Servers

In this article, we consider the dynamic allocation of bursty requests stochastically arriving at heterogeneous servers with uncertain setup times. Lower expected response time and less power consumption are desirable objectives of users and service providers respectively. However, sudden increase and decrease of cloud servers caused by bursty requests are rather challenging to get an appropriate trade-off between the two conflicting objectives which are closely related to the launched servers. The heterogeneity of the cloud servers further makes it more difficult to decide how to switch on and off servers and effectively and efficiently allocate bursty requests with balanced objectives. Based on a Markov decision process, a real-time bilevel decision-making model is constructed for unallocated requests which includes: whether to launch a server and which type of server to launch. A learn-and-deploy algorithm framework is proposed which contains two complementary stages. In the first stage, an effective offline bi-objective optimization algorithm is proposed to learn a set of policies, which provides helpful trade-off information for a decision-maker to choose a preferred policy <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">a posteriori</i> . In terms of the system status, a policy decides whether to launch a server according to a state-action table and which server to launch using a server priority sequence. In the second stage, a computationally efficient policy deployment method is proposed to search the corresponding action in the selected policy based on the current system status and apply it to the real-time system. Experimental studies over a large number of random and real instances have been conducted to validate the effectiveness of the proposed bilevel model and algorithm. Compared to the most recent existing method, the performance of the proposed approach can at most achieve an 80% improvement on power consumption and 20% improvement on response time.

노인장기요양서비스 준시장화와 공급관리체계의 변화

As a result of the promotion of quasi-marketization in the flow of welfare reform that reorganized the welfare state regime since the 1990s in various countries, the role of the government as a provider and financier has been reduced in the provision of welfare services, but the role of the government as a manager surrounding the provision of services in the quasi-market has been reinforced. However, the role of the government as a service provision management entity in the quasi-market has been deployed differently in different countries, and as a result of the policy deployment, it is noteworthy that the governance patterns following the promotion of quasi-marketization appear differently.&#x0D; Against this background, this study is intended to look at changes in policies surrounding the overhaul of the service provision management system, focusing on policy deployment for the overhaul of the provision management system following the quasi-marketization of long-term care service for the elderly, to see how such policies have been deployed in advanced countries with examples from representative countries (the UK and Japan) that executed the reform of quasi-marketization. Therefore, in order to determine the changes in the provision management system that has been overhauled following the quasi-marketization reform of long-term care service for the elderly in the UK and Japan since the 1990s, this study approaches the provision management system from the viewpoint of the “pluralization of provision management” to analyze how the management system (regulation and evaluation system) surrounding service provision in the quasi-market has been pluralistically overhauled. In particular, in relation to the pluralistic overhaul of the service provision management system, this study analyzes what roles are played by the government including local governments as managers after the establishment of the regulation and evaluation system to guarantee the quality of service following quasi-marketization and what relationships have been established with the public organizations that have newly emerged as regulation and evaluation bodies. Furthermore, by identifying changes in the role of local governments following the promotion of quasi-marketization through such policy deployment along with how the new governance frameworks for regulation, evaluation, devolution, etc., have been established and changed, this study captures the current state of pluralistic governance for long-term care for the elderly in the UK and Japan and presents implications for long-term care governance for the elderly in South Korea.

Race and the Use of Force by Police Revisited: Post-Ferguson Findings From a Large County Police Agency

This study examines racial and ethnic disparities in the use of force by the Fairfax County, VA Police Department using post-Ferguson data and after the adoption of modern training and use of force policies on de-escalation. White civilians were overrepresented as subjects of force compared to a variety of benchmarks, while Hispanics were underrepresented. Black and Asian civilians also were overrepresented using some benchmarks but not others. Multivariate models revealed that Black civilians experienced more severe force, higher average force levels, and more total force than Whites. We discuss possible reasons for the overrepresentation of Blacks and Whites as subjects of force, as well as the disparities in force levels and the totality of force experienced by Black civilians. We conclude with suggestions for policy change, training, and deployment changes to help reduce the potential influence of unconscious bias in the application of force by the FCPD and other law enforcement agencies.

Comparative Study on International Research Hotspots and National-Level Policy Keywords of Dynamic Disaster Monitoring and Early Warning in China (2000-2021).

For more than 20 years, disaster dynamic monitoring and early warning have achieved orderly and sustainable development in China, forming a systematic academic research system and top-down policy design, which are inseparable from the research of China's scientific community and the promotion of government departments. In the past, most of the research on dynamic disaster monitoring and early warning focused on specific research in a certain field, scene, and discipline, while a few studies focused on research review or policy analysis, and few studies combined macro and meso research reviews in academia with national policy analysis for comparative analysis. It is necessary and urgent to explore the interaction between scholars' research and policy deployment, which can bring theoretical contributions and policy references to the top-down design, implementation promotion, and academic research of China's dynamic disaster monitoring and early warning. Based on 608 international research articles on dynamic disaster monitoring and early warning published by Chinese scholars from 2000-2021 and 187 national policy documents published during this period, this paper conducts a comparative analysis between the knowledge maps of international research hotspots and the co-occurrence maps of policy keywords on dynamic disaster monitoring and early warning. The research shows that in the stage of initial development (2000-2007), international research articles are few and focused, and research hotspots are somewhat alienated from policy keywords. In the stage of rising development (2008-2015), after the Wenchuan earthquake, research hotspots are closely related to policy keywords, mainly in the fields of geology, engineering disasters, meteorological disasters, natural disasters, etc. Meanwhile, research hotspots also focus on cutting-edge technologies and theories, while national-level policy keywords focus more on overall governance and macro promotion, but the two are gradually closely integrated. In the stage of rapid development (2016-2021), with the continuous attention and policy promotion of the national government, the establishment of the Ministry of Emergency Management, and the gradual establishment and improvement of the disaster early warning and monitoring system, research hotspots and policy keywords are integrated and overlapped with each other, realizing the organic linkage and mutual promotion between academic research and political deployment. The motivation, innovation, integration, and transformation of dynamic disaster monitoring and early warning are promoted by both policy and academic research. The institutions that issue policies at the national level include the State Council and relevant departments, the Ministry of Emergency Management, the Ministry of Water Resources, and other national ministries and commissions. The leading affiliated institutions of scholars' international research include China University of Mining and Technology, Chinese Academy of Sciences, Wuhan University, Shandong University of Science and Technology, and other institutions. The disciplines involved are mainly multidisciplinary geosciences, environmental sciences, electrical and electronic engineering, remote sensing, etc. It is worth noting that in the past two to three years, research and policies focusing on COVID-19, public health, epidemic prevention, environmental governance, and emergency management have gradually increased.

Where Reinforcement Learning Meets Process Control: Review and Guidelines

This paper presents a literature review of reinforcement learning (RL) and its applications to process control and optimization. These applications were evaluated from a new perspective on simulation-based offline training and process demonstrations, policy deployment with transfer learning (TL) and the challenges of integrating it by proposing a feasible approach to online process control. The study elucidates how learning from demonstrations can be accomplished through imitation learning (IL) and reinforcement learning, and presents a hyperparameter-optimization framework to obtain a feasible algorithm and deep neural network (DNN). The study details a batch process control experiment using the deep-deterministic-policy-gradient (DDPG) algorithm modified with adversarial imitation learning.

How to improve energy efficiency policies to address energy poverty? Literature and stakeholder insights for private rented housing in Europe

Households in the private rented sector (i.e., households renting properties at a market rate from a private landlord), and especially those of lower income, face far more energy poverty issues than the general population, meaning that they are often unable to afford socially and materially required levels of energy services. In this context, energy efficiency policies are of the outmost importance as they can improve households' living conditions and contribute to the alleviation of energy poverty. Nevertheless, when it comes to the private rented sector across Europe, implementing such policies is hindered by several barriers. Our article sheds light on structural factors that serve as barriers to, and solutions for, the implementation of energy efficiency policies in the private rented sector across Europe. Our approach is twofold: first, an in-depth desk research is performed to identify financial, social, political/regulatory, and technical barriers and solutions. Literature findings are then supplemented and validated by a European-wide survey, eliciting knowledge embedded in 64 relevant stakeholders and field experts. Our findings indicate that although financial barriers are always of the essence, a holistic assessment of barriers is required, as there is a lack of awareness about policies that could improve tenants' living conditions. They also provide possible solutions that governmental bodies at all levels and other organisations could implement to improve energy efficiency and tackle energy poverty. Overall, our work suggests key implications, which, if acted upon, could accelerate the deployment of energy efficiency policies in the private rented sector across Europe.

Experts versus the Public: Perceptions of Siting Wind Turbines and Performance Concerns

Experiences of wind turbines (WT) shape public perception and acceptance of the technology, influencing government policy, deployment, and land-use policies of wind turbines. This paper attempts to find changes in public perceptions over the last three decades and differences between experts and the public over different land-use options. A semi-structured questionnaire that integrates a visual survey of 10 images of WT technology in different urban, landscape and seascape settings was presented to both groups. The perceptions of siting, proximity, landscape type, and maturity of urban wind turbines’ technology in renewable energy generation were contrasted. The results revealed that both the public and experts alike significantly preferred images of WT inclusion in seascape and landscape settings and responded negatively to images of WT as an addition to buildings in urban contexts. Images of wind turbines around transport settings were ranked in the second set of acceptances, after landscape settings, indicating that closer proximity to WT is acceptable, but for a short duration. The analysis also highlighted a preference by the public for aesthetically engaging WT, even if they resulted in lower energy yields, but were less accepted by the experts who based their judgment on technical performance.

A Transformer-Based Machine Learning Approach for Sustainable E-Waste Management: A Comparative Policy Analysis between the Swiss and Canadian Systems

Efficient e-waste management is crucial to successfully achieve sustainable urban growth universally. The upsurge in e-waste has resulted in countries, including Canada, adopting a wide array of policies associated with sustainable management. In this study, we conducted a mixed-method analysis of Canadian e-waste management policies to showcase the opportunities and limitations of the current system. We examine and compare the effectiveness of electronic waste management strategies in Canada and Switzerland using a comparative policy evaluation and by quantitatively measuring their efficiencies through two efficiency methods, namely a transformer-based, bidirectional, unsupervised machine learning model for natural language processing (NLP) and data envelopment analysis (DEA). Switzerland is utilized as a comparison case due to its robust legal framework that has been in place for proper management e-waste in order to enhance Canada’s electronic waste management system. The policy considerations presented in this study are directed toward urban planners, policy makers, and corporate strategists. These involve a mix of political, economic, social, and environmental planning tools concerning how to communicate and foster competent e-waste management in these countries. This is the first study to incorporate DEA and NLP-based BERT analysis to identify the most efficient policy deployment concerning e-waste management.

Reliable Off-Policy Evaluation for Reinforcement Learning

Off-policy evaluation is an important topic in reinforcement learning, which estimates the expected cumulative reward of a target policy using logged trajectory data generated from a different behavior policy, without execution of the target policy. It is imperative to quantify the uncertainty of the off-policy estimate before deployment of the target policy. Here we leverage methodologies from (Wasserstein) distributionally robust optimization to provide robust and optimistic cumulative reward estimates. With proper selection of the size of the distributional uncertainty set, these estimates serve as confidence bounds with nonasymptotic and asymptotic guarantees under stochastic or adversarial environments. We also generalize those results to batch reinforcement learning.

Competitive exclusion of two viral strains of COVID-19.

The pandemic COVID-19 has caused severe losses in public health and economy. One of the most difficult problems in prevention of the disease spread is the emergence of new variants. In this paper, a mathematical model is formulated, which captures the main feature of COVID-19 spread with two viral strains. It is shown by analytical method that the model exhibits the competitive exclusion principle, where one viral strain with the larger basic reproduction number is dominant and the viral strain with the smaller reproduction number is excluded. The results are important for the deployment of prevention policy of COVID-19.

Protests as a form of electioneering? A comparative study of the 2016 and 2021 local government elections protests in Gauteng and Kwa-Zulu Natal

ABSTRACT South Africa is gripped by protests which have usually taken the form of service delivery protests that have become more manifest during election periods. However, each election cycle has different types of protests with varying degrees of intensity. Thus, the aim of this paper is to analyse the link between elections and service delivery protests in South Africa with a focus on the 2016 and 2021 local government elections in Gauteng and KwaZulu-Natal provinces. This qualitative study relies on relative deprivation theory to tease out socio-economic issues driving these protests and rational choice theory to explore the rationale behind the timing of protests. Findings are that, in both provinces, although protests were linked to poor service provision, other factors like intra-party tension were notable. This article recommends a re-visit of cadre deployment policy, because the lack of capacitated and skilled cadres has fuelled service delivery especially in local municipality.

Edge-Based Optimal Routing in SDN-Enabled Industrial Internet of Things

The efficiency of communication between devices in any network depends on the network architecture and the routing algorithms. This efficiency vastly affects the productivity of the Internet of Things (IoT) nodes when utilized in an industrial environment. The conundrum of using the conventional network architecture in an industry/factory connected by IoT devices has paved the way for the development of software-defined networking (SDN). Traditional network architectures lack network programmability and hinder the testing and performance evaluation of new communication protocols. The use of vendor-specific hardware is a major hindrance to create a network where the control is centralized for the deployment of different policies over the network. This article aims to find an optimal path for efficient communication between the nodes present in a programmable Industrial IoT (IIoT) network. A programmable IIoT network is created and end-user routing algorithms are designed to find the optimal path for communication. We integrate an SDN framework with IIoT nodes for the creation of a centralized programmable network architecture and deploy custom routing algorithms over it for evaluation. To further reduce the communication latency, we incorporate edge computing servers (ECSs) in the network. Three routing algorithms: 1) <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> -means-based optimal path algorithm; 2) cluster-based optimal path algorithm; and 3) minimum interval-based optimal path algorithm, that can be deployed over an SDN are proposed in this article. Experimental results reveal that the proposed algorithms are more efficient than the existing algorithms.

Possibility of Sustainable Entry into Overseas Port Operation Markets by Japanese Companies

The Japanese government has set an overseas infrastructure deployment policy to involve Japanese companies in all upstream to downstream processes, that is, from project formation, procurement, and construction to operation and maintenance. Although Japanese companies have been hitherto involved in overseas port development through Official Development Assistance, their entry into overseas port operation projects has been limited, meaning the realization of the policy is not expected to be easy. This study thus examines the possibility of sustainable entry into overseas port operation markets by Japanese companies. Specifically, we review port governance in Japan, characteristics of the Japanese market in the global market, and the status of the participation of domestic terminal operators in Japan and overseas, and then identify the possibilities and methods of sustainable inclusion in the future overseas port operations by interviewing representative companies and the government. We finally provide future directions in terms of both increasing entry opportunities and improving the entry environment.

Weaving an innovation network from the middle-out: the case of the renewable energy ecosystem

BackgroundRenewable energy (RE) systems are becoming a central component of the clean energy transition and are often seen as the way to combat climate change. Their establishment requires innovation, investments, and deployment policies for emerging technologies. Governments around the world are increasingly trying to create and support the energy-tech and climate-tech innovation ecosystems in their attempt to promote an innovation-supporting environment. However, energy innovation policies are often aligned with the dichotomous notion of technology-push and market-pull and overlook the social, political, and cultural contexts in which RE innovative technologies are embedded, and the multiple actors and interactions that are associated with their development. By combining the Middle-Out Perspective (MOP) and innovation literature, this paper argues that an innovation ecosystem could be weaved from the middle-out.MethodsThis paper analyzes the case of ‘Eilat-Eilot Renewable Energy Initiative’ and Israel’s RE innovation ecosystem creation through the lens of the MOP and applies a socio-technical interpretation to the push and pull terminology. Process tracing methodology was applied to trace a sequence of events to determine whether an actor was pushed or pulled, the formation of a network of actors, and its evolvement to an energy innovation ecosystem from the middle-out. Data for the analysis were collected from interviews, policy papers, media articles, and Governmental decisions.ResultsThe analysis demonstrates how middle actors push the implementation of RE technologies in Israel, and at the same time pull decision-makers and other middle actors to promote the low carbon transition. The push and pull forces and the interactions between actors lead to the engagement of new stakeholders in the innovation network, the adoption of more ambitious RE targets and supporting policies, and the creation of an effective RE innovation ecosystem.ConclusionsThis paper uses the MOP as an analytical framework and the push and pull terminology to demonstrate how a middle actor initiates and develops an actor-network by interacting with other actors. As this network broadens, it forms an effective innovation ecosystem. A network of actors has the potential to lead change, provide innovative ideas, initiate research, encourage investments, create employment possibilities, and promote regional sustainable development.

Research and application of next-generation firewall technique in medical network

With the rapid development of medical informatization as well as the larger quantities of information and higher integration level, it has become a severe challenge to keep the core data confidential. In this thesis, the role of next-generation firewall in the medical network has been first introduced. Then, a medical network has been designed and an overall network building scheme aimed at the medical network has been planned based on the actual demand of a hospital. Then, VLAN has been divided and IP address has been planned to each floor. Moreover, relevant protocols and important equipment to be used have been introduced in detail, including DHCP protocol, SNMP protocol, next-generation firewall safety devices. In addition, the IP address of each area has been planned, the connectivity of LAN has been tested and the simulated attack and defense test has been taken based on the deployment principle of next-generation firewall. Furthermore, a detailed introduction of deployment of security policy as well as some common security vulnerabilities on firewall has been made. At last, the attack and defense test of firewall has been taken to prove the efficient protection function of firewall in the medical network and meanwhile prove that next-generation firewall technique could help the hospital solve the network security issues to some extent.

Estimation of Ebola's spillover infection exposure in Sierra Leone based on sociodemographic and economic factors.

Zoonotic diseases spread through pathogens-infected animal carriers. In the case of Ebola Virus Disease (EVD), evidence supports that the main carriers are fruit bats and non-human primates. Further, EVD spread is a multi-factorial problem that depends on sociodemographic and economic (SDE) factors. Here we inquire into this phenomenon and aim at determining, quantitatively, the Ebola spillover infection exposure map and try to link it to SDE factors. To that end, we designed and conducted a survey in Sierra Leone and implement a pipeline to analyze data using regression and machine learning techniques. Our methodology is able (1) to identify the features that are best predictors of an individual's tendency to partake in behaviors that can expose them to Ebola infection, (2) to develop a predictive model about the spillover risk statistics that can be calibrated for different regions and future times, and (3) to compute a spillover exposure map for Sierra Leone. Our results and conclusions are relevant to identify the regions in Sierra Leone at risk of EVD spillover and, consequently, to design and implement policies for an effective deployment of resources (e.g., drug supplies) and other preventative measures (e.g., educational campaigns).

Harmonizing the Electricity Markets in Africa: An Overview of the Continental Policy and Institutional Framework towards the African Single Electricity Market

Africa is a huge continent with an area equal to 30,244,049 km2 and a population of over 1.3 bn. According to the World Energy Outlook 2021, the electrification rate for Sub-Saharan Africa is at 79% for urban areas and only 28% for the rural areas. Given the low electricity infrastructure in Africa in production, transmission and distribution as well as the very low access in rural areas, it is understood that coordinated activities must take place for both on-grid and off-grid electrification activities, if the targets set for 2030 are to be met. This paper aims to present the challenges faced in the harmonization of the electricity markets in Africa due to the complexities of political continental integration, as well as continental economic integration, by performing a review of the progress made so far. It is one of the few efforts that have aimed to present in a single document the institutional framework of the electricity sector in Africa, and how these institutions collaborate in order to form and deploy policies at the continental and regional levels, affecting ultimately the deployment of policies at the national level. It also presents the current status of the continental electricity market policy framework activities towards the achievement of the 2030 goals in line with Agenda 2063, and the crucial role the African Single Electricity Market (AfSEM) is set to play. Finally, it provides lessons learnt and recommendations on facilitating the way forward in terms of the institutional actors’ collaboration, both for Africa and globally, in terms of developing regional energy markets.

DeepAir: Deep Reinforcement Learning for Adaptive Intrusion Response in Software-Defined Networks

In this paper, we propose an adaptive intrusion response solution based on deep reinforcement learning, namely DeepAir, to effectively defend against cyber-attacks in Software-Defined Networks (SDN). Specifically, we first study an intrusion response system (IRS) that operates at the SDN control plane. Next, we propose a dynamic intrusion response solution to maximize the attack defense performance while minimizing the negative impact on benign traffic forwarding and the policy deployment cost in the SDN data plane. Then, we model the intrusion response system based on a Markov decision process (MDP) approach and formulate the related optimization problem. Afterward, we develop a Double Deep <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">${Q}$ </tex-math></inline-formula> -Network based intrusion response control algorithm to assist the intrusion response system to quickly obtain the optimal intrusion response policy. In our case study, we consider denial-of-service (DoS) attacks—the performance evaluation results demonstrate that DeepAir can effectively prevent malicious packets from arriving at the victim in all considered DoS attack scenarios, i.e., approximately 85% of attack packets are dropped. Moreover, by applying the optimal intrusion response policy, DeepAir can significantly reduce the ratio of Quality-of-Service violated traffic flows compared to a <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">${Q}$ </tex-math></inline-formula> -learning based approach (by 70%), and to two existing solutions, i.e., GATE (by 75%) and GTAC-IRS (by 80%), respectively.

Forecasting the development trend of new energy vehicles in China by an optimized fractional discrete grey power model

As an effective technology to reduce traffic pollution emissions, the new energy vehicle industry has developed rapidly in recent years, and the sales of new energy vehicles have doubled in 2021. Accurately forecasting methods provide important references for industrial policy deployment, infrastructure construction and energy demand estimation. In order to predict the development trend of China's new energy vehicle industry under the limited data samples, this paper proposes an optimized discrete grey power model to fit the nonlinear relationship between grey information factor and temporal factor. Firstly, the grey differential equation and its discrete form are directly constructed by using power accumulated data, which simplifies the integral solution process of the traditional grey power model. Secondly, fractional accumulation operator is introduced into the discrete model to ensure the new information priority of the original data. Then, the heuristic algorithm is used to accurately estimate the new parameters of the proposed model. Thirdly, the proposed discrete model is the unified form of two existing discrete grey power models, which expands the scope of modeling and has higher adaptability. Simulation experiments and two numerical cases are used to verify the effectiveness of the proposed method. Finally, the proposed method is used to predict the annual sales and ownership of new energy vehicles in China. The predicted results show that by 2025, China's new energy vehicle sales will reach 8.84 million, accounting for about 24% of the total vehicle sales, surpassing the industry development target (20%) set by the Chinese government.