Search results

The purpose of this work is to bridge FL and blockchain technology through designing a blockchain-based smart agent system architecture and applying in FL. and blockchain technology through designing a blockchain-based smart agent system architecture and applying in FL. FL is an emerging collaborative machine learning technique that trains a model across multiple devices or servers holding private data samples without exchanging their data. The locally trained results are aggregated by a centralized server in a privacy-preserving way. However, there is an assumption where the centralized server is trustworthy, which is impractical. Fortunately, blockchain technology has opened a new era of data exchange among trustless strangers because of its decentralized architecture and cryptography-supported techniques.

In this study, the author proposes a novel design of a smart agent inspired by the smart contract concept. Specifically, based on the proposed smart agent, a fully decentralized, privacy-preserving and fair deep learning blockchain-FL framework is designed, where the agent network is consistent with the blockchain network and each smart agent is a participant in the FL task. During the whole training process, both the data and the model are not at the risk of leakage.

A demonstration of the proposed architecture is designed to train a neural network. Finally, the implementation of the proposed architecture is conducted in the Ethereum development, showing the effectiveness and applicability of the design.

The author aims to investigate the feasibility and practicality of linking the three areas together, namely, multi-agent system, FL and blockchain. A blockchain-FL framework, which is based on a smart agent system, has been proposed. The author has made several contributions to the state-of-the-art. First of all, a concrete design of a smart agent model is proposed, inspired by the smart contract concept in blockchain. The smart agent is autonomous and is able to disseminate, verify the information and execute the supported protocols. Based on the proposed smart agent model, a new architecture composed by these agents is formed, which is a blockchain network. Then, a fully decentralized, privacy-preserving and smart agent blockchain-FL framework has been proposed, where a smart agent acts as both a peer in a blockchain network and a participant in a FL task at the same time. Finally, a demonstration to train an artificial neural network is implemented to prove the effectiveness of the proposed framework.

Smart spaces are open complex computing systems, consisting of a large variety of cooperative smart things. Central to building smart spaces is the support for sophisticated coordination among diverse smart things collaborating to accomplish specified tasks. Multi‐agent systems are often used as the software infrastructures to address the coordination issue in smart spaces. However, since agents in smart spaces are dynamic, resource‐bounded and have complicated service dependencies, current approaches to coordination in multi‐agent systems encounter new challenges when applied in smart spaces. The purpose of this paper is to address these issues.

The paper presents Baton, a service management system to explicitly resolve the particular issues stemming from smart spaces when coordinating agents. Baton is designed as a complement to coordination approaches in multi‐agent systems with a focus on mechanisms for service discovery, composition, request arbitration and dependency maintenance. Baton is now deployed in our own smart spaces to achieve better agent coordination.

The effectiveness and efficiency of Baton is validated by its practical use in the designed scenario and some evaluation experiments.

An attempt at performing dynamic service composition in Baton is made by using semantic information in future work.

Baton, a service management system to explicitly resolve the particular issues stemming from smart spaces when coordinating agents is presented.

The smart open house provides optimal adaptability using sensing, operating, information, and communications technology, in conjunction with open building in-filled components, to perceive user needs and environmental changes, and thereby meet the needs for sustainability and a healthy living environment. These needs are particularly pressing in view of the aged society that will emerge in Taiwan after 2020. Based on the smart open house hypothesis, this study proposes using agent-based smart skins in a smart open house, where an agent-based smart skin is embedded in a lifetime home (or ageless home) with an open system construction. The agent-based smart skin operating mechanism employs fuzzy logic inference and neuro-fuzzy learning to process environmental information from sensing devices and drive skin elements, achieving adaptive action, meeting residents' lifetime use needs, and offering a user experience-oriented smart care capability.

Cities evolved into quite complex urban systems. The rigid management process must reflect the complexity of the current political, social, and economic environment. With the vast city growth, citizens experience new difficulties – traffic congestion, pollution, immigration, overcrowding, and inadequate services.

In our research, we analyze problems and benefits that occur with the growing complexity and offer a new concept considering every city as a live and constantly developing complex adaptive system of many participants and actors that operate in an uncertain environment. These actors (residents, businesses, transport, energy, water supply providers, entertainment, and others) are the main elements of city life.

The new concept of “Smart City 5.0” is based on a previously developed model of Smart City 4.0 (compared with Industry 4.0) and implements the Urban Digital Ecosystem, where every element can be represented by a smart agent operating on its behalf. It is shown that smart services can interact vertically and horizontally in the proposed ecosystem, supporting competition and cooperation behavior based on specialized network protocols for balancing the conflicting interests of different city actors.

The chapter describes the design principles and the general architecture of the Urban Digital Ecosystem, including the basic agent of smart service, protocols of the agent’s negotiation, the architecture, and basic principles Smart City knowledge base.

The developed evolutionary methodology of implementation will ensure a minimum of disruptions to city services during its transformation into an urban ecosystem to harmoniously balance all spheres of life and the contradictory interests of different city actors.

In e‐commerce, trust becomes an essential prerequisite for customer relationship building. Drawn from established theoretical work on trust and relationship marketing, a model is proposed aiming to help in highlighting the differences between traditional and e‐commerce and to facilitate thinking as to how trust can be built in virtual environments. Conceptualized in the context of an electronic servicescape, the model helps to demonstrate how agent and virtual reality technologies can facilitate the expressiveness required for the formation of trust through iterative interaction with promises being made, enabled and fulfilled.

The purpose of this paper is to develop an intelligent manufacturing system for transforming production management and operations to an Industry 4.0 manufacturing paradigm.

A manufacturing mode-Graduation Manufacturing System is designed for organizing and controlling production operations. An Industrial Internet of Things (IIoT) and digital twin-enabled Graduation Intelligent Manufacturing System (GiMS) with real-time task allocation and execution mechanisms is proposed to achieve real-time information sharing and production planning, scheduling, execution and control with reduced complexity and uncertainty.

The implementation of GiMS in an industrial company illustrates the potential advantages for real-time production planning, scheduling, execution and control with reduced complexity and uncertainty. For production managers and onsite operators, effective tools, such as cloud services integrates effective production and operations management strategies are needed to facilitate their decision-making and daily operations at the operational level.

This paper presents an Industry 4.0 paradigm-GiMS, which aims to explore Industry 4.0 technologies opportunities on operations and production management, especially on production planning, scheduling, execution and control.

This paper aims to study the relationship between personal knowledge management (PKM) and organisational knowledge management (OKM).

Based on the proposed framework of PKM and OKM, the author has analytically modelled and studied the strategies of linking the two to manage personal knowledge for organisational purposes.

Applying a framework categorising PKM and OKM from two dimensions – knowledge properties and management perspectives – the author proposes the strategies of personalising organisational knowledge (POK) and organisationalising personal knowledge (OPK) as the effective linkage between PKM and OKM. Integrating a knowledge management system (KMS) into the framework, the author determines the optimal level of KMS in facilitating the implementation of POK and OPK strategies.

Future research may relax some of the assumptions and empirically verify the analytical results.

The research provides guidelines for matching investments in information technology with appropriate knowledge management strategies.

With a unique approach, the paper contributes to the missing link between PKM and OKM.

The purpose of this paper is to critically analyze whether supply networks may be validly treated as complex adaptive systems (CAS). Finding this to be true, the paper turns into the latest concerns of complexity science like Pareto distributions to explain well‐known phenomena of extreme events in logistics, like the bullwhip effect. It aims to introduce a possible solution to handle these effects.

The method is a comparative analysis of current literature in the fields of logistics and complexity science. The discussion of CAS in supply networks is updated to include recent complexity research on power laws, non‐linear dynamics, extreme events, Pareto distribution, and long tails.

Based on recent findings of complexity science, the paper concludes that it is valid to call supply networks CAS. It then finds that supply networks are vulnerable to all the nonlinear and extreme dynamics found in CAS within the business world. These possible outcomes have to be considered in supply network management. It is found that the use of a neural network model could work to manage these new challenges.

Since, smart parts are the future of logistics systems, managers need to worry about the combination of human and smart parts, resulting design challenges, the learning effects of interacting smart parts, and possible exacerbation of the bullwhip effect. In doing so, the paper suggests several options concerning the design and management of supply networks.

The novel contribution of this paper lies in its analysis of supply networks from a new theoretical approach: complexity science, which the paper updates. It enhances and reflects on existing attempts in this field to describe supply networks as CAS through the comprehensive theoretical base of complexity science. More specifically, it suggests the likely vulnerability to extreme outcomes as the “parts” in supply networks become smarter. The paper also suggests different ways of using a neural network approach for their management – depending on how smart the logistics parts actually are.