End-product quality prediction is crucial in smart manufacturing, where reliable evaluation and parameter optimization are essential for ensuring high-quality outputs. This study presents a novel approach that combines adaptive machine learning and nonlinear regression to accurately predict the quality of highly customized end products using limited supply-chain data through digital transformation. The research was conducted in collaboration with a major power transformer manufacturer and its supply chain partners. The adaptive model was trained and validated using real datasets from key components provided by the supply chain, resulting in accurate predictions of end-product quality. The model incorporates the core loss parameter, obtained from the power transformer's key component, as an input dataset for training and testing. The proposed approach, called AdaBoost-Regression, combines adaptive boosting (AdaBoost) and Regression machine learning techniques. Experimental results demonstrate that the AdaBoost-Regression model outperforms simple AdaBoost and Regression models in predicting transformer quality. The model also exhibits superior performance in terms of mean absolute percentage error (MAPE) and root mean square error (RMSE) during real-data verification. This approach has the potential to significantly reduce overall production costs by accurately predicting the quality of complex, expensive, and highly customized industrial products. It can be applied across various industrial sectors.

Request for quotation (RFQ) is a lengthy document soliciting vendor products and services according to rigid specifications. This research develops an integrated natural language processing (NLP), text mining, and machine learning approach for intelligent RFQ summarization. Over 1,300 power transformer RFQ requests are used to build a word-embedding model for training and testing. Domain keywords are extracted using N-gram TF-IDF. The method automatically extracts essential specifications such as voltage, capacity, and impedance from RFQs using text analytics. The K-means algorithm groups the sentences of each specification. The TextRank algorithm identifies important sentences of all specifications to generate RFQ summaries. The summarization system helps engineers shorten the time to identify all specifications and reduces the risk of missing important requirements during manual RFQ reading. The system helps improve the complex product design for manufacturers and improve the cost estimation and competitiveness of quotations in a highly competitive marketplace.

The current virtual reality (VR) techniques develop immersive environments via inducing illusions to our sense. Nowadays, most of VR focuses on inducing visual illusion. Hence, visual is the most important input channel for experiencing and exploring the VR environments. Recently, extensive research efforts have been put on eye-tracking studies. However, the development and growing trends of the VR-based eye-tracking studies are unrevealed due to the lack of a systematic literature review on it. In this study, we reviewed related literature from 2000 to 2019 and summarized them into two main categories, including eye tracking methods and eye-tracking-enabled applications, such as tracking gaze points to manipulate the VR environment, measuring user states, and evaluating the usability of VR based on eye-tracking data. Based on the literature review, we can find that eye-tracking can assist in developing adaptive VR systems and enhance users experience. While comparing with 2D environments, immersive VR environment still requires more deep studies in eye-tracking.

Global online sales for products, where many are substantially identical or deceptively similar, are the cause of a growing number of trademark (TM) infringement lawsuits. This research proposes an intelligent trademark legal precedent recommendation system to assist trademark owners to find relevant past cases, laws, and judgments to form legal arguments to defend against infringement. Judicial precedent and applicable laws from the USA are used to construct an ontology of trademark litigation knowledge. The ontology is used to analyze potential infringement cases with similar laws and precedents used to resolve previous legal disputes. The analysis provides a basis for proceeding with legal action necessary to protect a company's brand equity when arguing potential trademark infringement. Using the Python programming language, the precedent-based recommendation system provides a means for continuously updating trademark case data and assists TM owners to quickly identify similar cases to support infringement allegations.

E-commerce provides a global platform supporting product transactions through the consumer purchase lifecycle including communications of perceived satisfaction and dissatisfaction. The customer feedback functions and social networks of many e-commerce websites allow for the creation of extremely large databases that can be mined to model the customers' perceptions toward online purchases. This research uses online customer reviews as the business intelligence corpus to help companies redesign products that better satisfy consumer preferences and differentiate their product offerings. After identifying the specific webpages of customer reviews, a web crawler collects review text. Computer-supported text mining, cluster analysis, and perceptual mapping are combined as a systematic analytic approach to compare products in a given domain. The study assists phone manufacturers to understand the positive and negative perceptions of customers related to their post-purchase experiences. The customer-preferred product functions, features, and price positions provide valuable strategic intelligence for new product designs and market differentiation.

The growth of global patents increased over the last decade as enterprises and inventors sought greater protection of their intellectual property (IP) rights. Global patents represent state-of-the-art knowledge for given domains. This research develops a hierarchical Latent Dirichlet Allocation (LDA)-based approach as a computational intelligent method to discover topics and form a top-down ontology, a semantic schema, representing the collective patent knowledge. To validate the knowledge extraction, 1,546 smart retailing patents collected from the Derwent Innovation platform from 2011 and 2016 are used to build the domain ontology schema. The patent set focuses on in-use, globally established, and non-disputed IP covering payment, user experience, and information integration for smart retailing. The clustering and LDA-based ontology system automatically build the knowledge map, which identifies the technology trends and the technology gaps enabling the development of competitive R&D and management strategies.

Conversational intelligence and the rise of its agent through “chatbots” are disrupting businesses worldwide. Chatbots enable natural interactions via text or voice leveraging technologies. Chatbots are used for various practical purposes such as information acquisition, customer service, and virtual assistance. Google Assistant, Apple Siri, and Amazon Alexa are some of the most widely used virtual assistants. Further, several use cases demonstrate a customer preference for chatbot interface against conventional graphical user interfaces. This has led to rapid growth in chatbot research and consecutively has led to a growth in the filing of patents. The disclosed patents represent major advances in the technology and offer hints about the directions conversational chatbots will grow in the future. Businesses wanting to integrate chatbot technology must, therefore, conduct an exhaustive patent analysis to identify required technologies, functions, and their associated product applications. This research presents a text mining based Technology Function Product Matrix (TFPM) generation approach to analyze a corpus of more than 400 chatbot technology patent documents. The analysis systematically identifies hotspot distributions across the patent landscape with respect to the key technologies, functional advances, and novel applications as current developments in chatbot technologies.

Computer vision (CV) attempts to mimic human eyes for image processing and identifications of detailed visual information, such as object positions, features of appearances, and even human emotions and behaviors. In this research, more than one hundred literatures, relating to applying deep learning (DL) methodologies in advanced computer visions (2010~2018), are reviewed and analyzed. The objective is to discover the state-of-the-art DL methods, topics, and trends for CV and their practical applications. DL algorithms aim at representing multi-levels of distributed neural networks. Because of the enhancement of high speed computational power, DL modeling, based on accumulated big data analytics, has found practical applications for non-supervised intelligent decision supports, such as detection of product defects and prognosis of machine malfunctions based on real-time signal or feature data analyses. There are a vast number of literature, describing DL related researches, developments, and implementations for problem solving. For the comprehensive mining of the related literature, we integrate Latent Dirichlet Allocation (LDA), K-means (Clustering), and normalized term frequency-inverse document frequency (NTF-IDF) approaches to discover, or called technology mining, of the major trends in DL for computer visions, specifically for key applications in object detection, semantic segmentation, image retrieval, and human pose estimation.

An ontology-based patent roadmap and valuation analysis approach provides competitive foresights for a corporation's associated patent portfolios and the underlying business processes. To demonstrate the methodology and computer supported analysis system, the patent portfolios of two global smart e-retailers are compared. The patents in each company's portfolio are text mined and categorized based on the smart retailing ontology schema. Both cases' text mining results are plotted as patent portfolio roadmaps, linking patents to innovation categories, commercial applications, and business process models. The three-dimensional (3-D) patent valuation analysis method provides foresights of the companies' competitive advantages related to “patent scope” (PS), “patent importance” (PI), and “patent innovation” (PIN). The research implements a computer supported system to provide evidence how companies utilize patent portfolios as key strategies for protecting business related intellectual properties (IPs) while implementing sustainable and competitive business processes.

Manufacturers are making a transition to intelligent production, particularly in high-tech industries. Robotic arms play a crucial role in the effectiveness of transforming processes. This research searches and analyzes patents related to robotic arms utilized by intelligent manufacturing systems. For this preliminary study, global patent trends are discussed which help evaluate the technology domain to be developed. Key words from a literature review are used to search for relevant patents, and then the patent text is analyzed using text mining software and program. The core technique applied to the patent document analysis is “Term Frequency-Inverse Document Frequency” (TF-IDF). Using TF-IDF, the patent documents are grouped into meaningful sub-technology cluster. Thus, the economic data related to the semiconductor industry is analyzed to correlate the economic impact which may be related to the transitions in technology. The data of Taiwan firms were targeted for anticipating research directions and expand intellectual property domains to invest in and pursue. For the advanced analysis, the international patent codes for global and local companies are compared. The final results help define the strength and weakness of Taiwanese robotic technology development for the semiconductor industry. Patent cluster and patent distribution analysis via a technology function matrix are the primary research approaches used for the case study. The research prepares for the development of a theoretical foundation to predict intelligent manufacturing trends.

With the growing population of digital users gaining online consumption power, retail industries are facing evolutionary changes in attracting and maintaining online and offline customers. The concepts and implementation of smart malls and retail shops are proposed by retailers or solution providers to satisfy new customer demands. The foundation of smart retailing includes information integration, customer experience, and e-transactions. This research investigates the economic impact of smart retailing and the growth of related intellectual properties (IPs), specifically, the retail domain online related patents. The technological aspect of smart retailing is organized by constructing a patent ontology map to define the domain boundary. The economic impact of technological transformations is studied by measuring the economic growth of the leading regions and countries and the representative patent assignees. Consumption preferences and labor structure changes in relation to the retail industry are analyzed to verify the economic impact of corresponding intellectual properties. The goal of the research is to develop regional development strategies for smart retailing using global patent and economic impact analyses.

Patent analysis helps companies understand their intellectual property (IP) portfolio and develop competitive marketing and management strategies. A Technology Function Matrix (TFM) is a critical approach for patent data analytics. This paper develops a generic computer supported TFM construction methodology that can be used for creating patent technical maps for any given domain. The approach is adopted for the case of the Internet of Things (IoT) patent technology analysis in the context of Industry 4.0 [1]. The aim of this article is to provide the methodology and analysis methods for IoT patent TFM and introduce computer supported IP and patent knowledge e-discovery.

Useful and innovative service design can increase organization competitiveness and consumer satisfaction. The problems facing consumers in the original open-shelf cosmetics service model includes the following issues: 1. Insufficient information on beauty goods for customers to make decisions, 2. Self-trial of products is needed but not easy to judge the quality of products, 3. It's hard to determine what's the reasonable price of the value of products, 4. To refund is full of trouble and difficulty. This study investigates the case of CSD company, applying the TRIZ (Theory of Inventive Problem Solving) to develop an innovative product-service system (PSS) for the brick cosmetic retail industry in order to enhance the customer experience and bring the crowd from online to offline. Analysis results find six TRIZ innovative principles, which are 10-prior action, 11-beforehand cushioning, 28-mechanics substitution, 32-color changes, 34-discarding and recovering, 35-parameter changes. These principles raise the conclusions and design implications for the case company.

With the rapid development of LTE-Advanced (LTE-A) mobile communication technology, companies continue to file for new patents or obtain licenses from patent assignees to secure the right to use LTE-A related intellectual properties (IPs) for product commercialization. There are many standard essential patents (SEPs) complying with LTE-A specifications set by the third generation partnership project (3GPP). LTE-A patent pools are collaborative efforts to gather SEPs from IP owners to reduce the process time of SEP licensing for manufacturers. This research develops the computer supported analysis process for the comparative study of two major LTE-A patent pools (i.e., Sisvel and Via Licensing). The first step is to search the ETSI database for LTE-A SEPs. The second step analyzes the patents in both pools based on IPC classification and LTE-A ontology definition. The main goal of the research is to reveal the patent profiles of LTE-A patent pools for collaborative and collective SEP licensing, which enables LTE-A technology developers and adopters (such as mobile device manufacturers) to fully cooperate in new product development and commercialization efficiently and economically.

The effective service design can be achieved by implementing collaboration work from domain experts and customers. This research depicts a conceptual service design framework: design-oriented systematic inventive thinking (DSIT) which can be applied in different problem contexts. It divides the real world design scenarios into four dimensions, and then provides the fitting integrated service design approach based on TRIZ. Four types of service design approaches have been suggested, conceptualized as “Human-independent service engineering”, “Problem-clarified service engineering”, “Solution-converged service engineering” and “Designing for services”. Regardless of any industries or companies, we can use this procedure to analyze and develop the suitable customized service design strategy for them.

Electronic payment (e-payment) is a subset of an e-commerce (EC) and critical to enhance customer loyalty. A well-designed e-payment service creates new commercial advantage and should be protected as intellectual property. This research develops an ontology roadmap using computer assisted methods to classify e-payment patents. The results of roadmap provide patent portfolio visualization which helps EC enterprises utilize strategic information for competitive advantages. This research also employs patent value indicators to benchmark patent portfolios. In the case implementation, two leading EC enterprises, Amazon and Alibaba, are studied using their e-payment patents as input to implement the proposed methodology. The roadmaps and value assessment results help to describe their underlying competitive advantages and provide management implications for business development.

For competitive global markets, patents are useful for protecting intellectual property rights (IPRs) or defending against infringement threats from competitors. In the mobile communication industry, companies such as Qualcomm and Ericsson dominate the industry with essential patents. Some non-practicing entities (NPEs) seek opportunities to license their IPRs and often challenge infringement of their IPRs by manufacturers or channel distributors. In order to avoid NPE patent infringement litigation and understand the patterns of NPE legal actions, this research focuses on developing a computer supported intelligent method to analyze patent and IP infringement litigation and their evolution trends. We use modified formal concept analysis to explore the evolution of the patent infringement lawsuits and their disputed patents. Two modified FCA models are constructed to observe the evolution of the court cases and the disputed patents. The research provides companies with technical references based on the evolutions of litigation and patent trends for future strategic R&D directions.

Two approaches are frequently used to study the critical technology and market trends of industry. One analyzes the business processes from the market perspective with statistics market survey. The other perspective predicts new technologies from the patent (intellectual property) data. Business process analysis is limited to analyzing existing technologies and market applications. Since the service delivery processes are not frequently linked to the intellectual property of the underlying technologies, applying a patent perspective enables the analysis of technology trends, technology ownership, and royalty payment analysis. These data are used to illustrate the technology life cycles, predict emerging trends, and identify research opportunities with Computer Supported tools. Patent analysis enables the predictions of technology trends but poorly links the technology to actual products for market adoption cause of the large amount of the data. This research develops an innovative computer supported approach for technology analysis combining both market and patent perspectives. The business processes of a given knowledge domain, driven by the market perspective, is analyzed to build the basis of the domain's ontology. Afterward, the key phrases of the business processes are added to form the ontology schema. The related patents are then systematically linked to the business process schema nodes. The research approach enables the business process-oriented patent portfolios among current and potential competitors to be mapped and critically compared from both market performance and technology ownership perspectives. The methodology allows in-depth integrated views of rivals' patenting and business development for strategic R&D management. The case of InvisalignTM innovative orthodontic services and the patent portfolio is studied to demonstrate the proposed methodology.

The rapid growth of mobile device usage stimulates the application of high-speed mobile broadband services. Thus, the adoption of Long Term Evolution (LTE), the leading standard of the fourth generation (4G) of mobile phone communication technology, becomes the main growth strategy for carriers. However, the LTE network services do not transmit voice data until the Voice over LTE (VoLTE) initiatives. This study applies quality function deployment (QFD) to explore the customer requirements and identify prospective technologies of VoLTE services. Our research shows that the top three important VoLTE techniques are related to “improve efficiency and resource utilization”, “enhance the functionality of the handover selection function nodes” and “circuit switched service over a long term evolution network”. Furthermore, the study finds that VoLTE outperforms Over-the-top (OTT) services in most of the customer requirements, which assures that VoLTE has its competitive advantage when compares to OTT services in the mobile voice call services.

This research focuses on a methodology applying Internet of Things (IoT) technology and enabling the material procurement process improvement of a manufacturer. The one stop logistic service provider (1SLP) is an integrator that assembles the resources, capabilities, and technologies of supply chain networks to design and implement comprehensive logistic service solutions. This research develops the IoT enabled 1SLP process framework and uses a case example to demonstrate the improved process with a system dynamic (SD) assessment. The current material procurement process of the case requires excessive operation times and the related data do not integrate with the supply chain related information. The 1SLP offers to-be (improved) logistics support incorporating IoT to shorten the operational time and enhance the information flow accurately and transparently. For the SD assessment, the results show that the to-be model improves the IoT process efficiency with the IoT when benchmarked with the as-is current processes.

Welcome to the 2014 18th IEEE International Conference on Computer Supported Cooperative Work in Design (CSCWD 2014), May 21–23, 2014. With great pleasure, we welcome you to the beautiful campus of National Tsing Hua University (NTHU) in Hsinchu, the Science City of Taiwan.

Manufacturing industries are adjusting their operations and strategies to gain sustainable competitive advantage. Many enterprises regard logistics as the core competency of their corporate strategy. Global logistic service companies often report limitations and challenges to supply chain integration including decreased transportation efficiency, diminished information transparency, and reduced material, information and cash flows. In this research, the concept of one-stop logistic services is defined and developed to provide enterprises with improved, integrated, and comprehensive services. A multi-view process modeling approach is applied to construct the model and identify potential bottlenecks with existing logistics processes. The one-stop logistic management model is defined as a to-be improved logistic service framework with functions to enhance the flow of materials, products, information, and cash transactions for integrated supply chains.

When the majority of energy (>90%) is generated by fossil fuels, carbon dioxide emissions increase the greenhouse gas effect of the region. Therefore, renewable, sustainable, and economically viable energy sources are needed as alternatives to fossil fuels. The facilities and installation costs for generating renewable energy is much higher than the cost of fossil fuel facilities. Thus, governments need effective policies, regulations, and incentive programs to promote the usage of renewable energy. Renewable energy can be classified into different categories, such as offshore and onshore wind power, photovoltaic solar power, and geothermal generated power. The policies used for promoting specific categories vary significantly. These policies depend on the policy goals, regulations, taxation, incentives, and promotional schemes. The purpose of this study is to apply clustering techniques and the analytic hierarchy process (AHP) to analyze types of renewable energies and their attributes with respect to economic factors, energy resources and supplies, and their environmental effects. The AHP approach is used to evaluate actions that can resolve challenges found in the development of renewable energy. The study provides scientific results to help government managers plan renewable energy policies. The data for the case study are collected from Taiwan's renewable energy statistics related to photovoltaic cells, wind farms, ocean thermal energy, geothermal energy, hydro power, and solid waste fuels. The research has two major results and findings. First, analytic models are developed for the decision support of renewable energy policies using intelligent clustering techniques. Second, the most suitable policies for promoting four renewable energy clusters are identified using the AHP approach.

The availability of modern technologies depends on the intellectual property licensing transactions. Patent pools, in particular, have emerged as important tools for organizing licensing activities. Traditionally, a patent pool aggregates the patents necessarily to practice a technology and avoids the transactional cost of separately negotiating licenses with multiple patent holders. More recently, competitive patent pools have emerged whereby multiple patent pools directed to related technologies vie for high-quality and essential patents and then competes for licensees. This research develops an integrated patent quality matrix for assessing the competitive features among multiple competing patent pools. The proposed matrix is built based on three primary patent quality dimensions: investment, maintenance (competitiveness) and litigation. Each dimension is composed of well-verified indicators. Based on the matrix result, we can identify the features and distribution of technology life cycle, commercial worth, and legal status of pool patents. This kind of patent quality intelligence can benefit potential licensees in assessing preferable patent pool.

In promoting the renewable and sustainable energy installation, carbon taxation is one of the commonly applied policies in controlling the fossil fuel energy consumption and production, particularly in the type of fossil fueled energy. Carbon tax is levied based on the quantity of carbon dioxide generated during energy production and consumption. Therefore, how to decide a reasonable tax rate is a critical issue for governments when setting an executable tax bill. This research uses System Dynamics (SD) approach to evaluate the influence of carbon tax on the installation of PV systems and the effects on electricity costs. Meanwhile, the decision analysis of the taxation allocation on supporting renewable energy (in our case study, the photovoltaic) system is conducted. The SD qualitative and quantitative models are built based on several realistic scenarios using Taiwan's national data to evaluate the achievement and benefit of the carbon tax policy. These results are valuable references for government energy policy makers globally with respect to developing country-specific carbon tax framework and subsequently promoting photovoltaic systems as effective renewable energy solution.

Patent documents consist of well-structured descriptions of the technology innovation and the research results. Patent documents are typically written with specific terminologies which often require the expertise of domain experts to interpret and organize for patent analysis. The rapid growth of number of patents applications and registry increases the difficulty in extracting and synthesizing knowledge from patents. The aim of this paper is to construct a framework of expressing and extracting accurate knowledge from patent documents that are related to dental implant connections. First, domain related patents are collected from the United States Patent and Trademark Office. Text mining techniques are used to analyze the key words in the sample set of patents. An ontology is created to express the domain knowledge by using a new ontology engineering technique. In recent years, the aging population has increased the demand for new products and innovations in the dental implant industry. Dental implants are a global, medically accepted treatment used to restore the human masticatory function. The patents collected in the domain of implant connections are used to test the validity and reliability of the ontology development framework. The study of patents related to dental implant connections is considered a critical part of advancing dental implant devices since these designs and innovations are subjected to extensive clinical trials. Given an accurate ontology schema, dental researchers correlate successful prior art designs to successful clinical trials to better understand the interaction between a new artificial body part and the response of the surgical implantation. The value of a patent increases dramatically if the device can be safely used by dentists to treat patients with fewer failures and longer utility.

Manufacturers and other businesses are under increasing pressure to improve productivity while reducing environmental impact. An environmental management accounting approach, called Material Flow Cost Accounting (MFCA), was developed in Germany in late 1990s and, since then, was adopted widely in Japan and other countries. MFCA is a management information system specializing in tracing all input materials flowing through production processes and measuring outputs in finished goods and wastes. To standardize MFCA practices, working group (WG) 8 of ISO technical committee ISO/TC 207 has developed and officially announced ISO 14051 framework in 2011. InnoLux Corporation (InnoLux) is one of global manufacturers which first introduced MFCA in LCD and optoelectronic industry. InnoLux adopted MFCA in four collaborative factories located in Nanhai area, Guangdong Province, P.R. China. This research has two main goals. First, we develop the framework of ISO14051-complied MFCA information system. Second, the case study is carried out to analyze and compare the performance before and after MFCA implementation in InnoLux's Nanhai factories for InnoLux's pursuit of collaborative green manufacturing.

This study focuses on designing a comprehensive set of key processes for semiconductor manufacturing supply chain and evaluating the relative importance of these key processes. The design and evaluation are performed by a multidisciplinary team consists of over twenty members from both academia and industry. This study adopts focus group methodology and fuzzy analytic hierarchy process (FAHP) for collaboration. The design and evaluation are based on experiences of these research team members who joined a successful supply chain management (SCM) project between the largest semiconductor foundry in the world and the world's largest semiconductor testing and packaging service provider. The result of the design is a hierarchy consists of 4 dimensions and 15 key processes. The resulting weighting from FAHP analysis indicates that both of the highest-ranking 1 dimension and the highest-ranking 4 factors account for approximately half of the overall weighting in their level. The designed key process hierarchy can serve as a checklist; helping SCM project managers avoids costly failure by ensuring that all key processes are well supported. The weightings provide SCM executives with a reference for the relative importance of these key processes and can also help them make critical decisions.

Patents are critical intangible assets for enabling an enterprise's creation and technology. Patent knowledge management is a time consuming task that often dominates the valuable time of R&D staff. This research develops an intelligent recommendation methodology and system for efficient and effective patent search. The system provides an algorithm which is an automatic search engine with management modules. The system clusters users' patent search behavior and infers new patent recommendations. The proposed methodology evaluates the filtered information of the searched patents. Afterward, the system clusters the users and finds each user's neighbors based on the collaborative filtering mechanism. By clustering neighbors' behaviors, the proposed system recommends new users to more appropriate patents to study. When enterprises are planning R&D policies or searching patents for possible prior arts, the intelligent recommendation system helps identify and recommend comprehensive and related patents while saving time and costs.

The purpose of this study is to analyze and plan the logistics model for Special Economic and Trade Zones (SETZ). Firstly, the orientation and the characteristics of the different types of domestic special zone is compiled. The preferential laws of international SETZ are then summarized, with a discussion regarding the supplementary measure when setting the SETZ in order to encourage Taiwanese entrepreneurial investment or to improve foreign companies setting up headquarters in Taiwan. Lastly, the logistics model for SETZ's characteristics is planned. This study takes the manufacturing and the export of large-scale engineering assets as an example, analyzing the supply chain's status of large equipment. Afterwards, according to the case company's business strategy and demand in the future, plans the operation model as a foundation and a basis of building the SETZ when the case company entering and being stationed in SETZ.

Engineering asset management (EAM) is a broad discipline and the EAM functions and processes are characterized by its distributed nature. However, engineering asset nowadays mostly relies on self-maintained experiential rule bases and periodic maintenance, which is lacking a collaborative engineering approach. This research proposes a collaborative environment integrated by a service center with domain expertise such as diagnosis, prognosis, and asset operations. The collaborative maintenance chain combines asset operation sites, service center (i.e., maintenance operation coordinator), system provider, first tier collaborators, and maintenance part suppliers. Meanwhile, to realize the automation of communication and negotiation among organizations, multiagent system (MAS) technique is applied to enhance the entire service level. During the MAS design processes, this research combines Prometheus MAS modeling approach with Petri-net modeling methodology and unified modeling language to visualize and rationalize the design processes of MAS. The major contributions of this research include developing a Petri-net enabled Prometheus MAS modeling methodology and constructing a collaborative agent-based maintenance chain framework for integrated EAM.

Nowadays many companies rely on patent engineers to search patent documents and offer recommendation and advice to R&D engineers. Given the great number of patent documents, new means to effectively and efficiently identify and manage the technology-specific patent documents are required. This research applies back-propagation artificial neural network (BPANN), a hierarchical ontology, and Normalized term frequency (NTF) method for binary document classification and content analysis. This approach helps to minimize inappropriate patent document classification. Hence, the approach reduces the effort to search and select patents for analysis. Finally, this paper use the design of exposure machines as a case study to illustrate and verify the efficacy of the approach proposed in this paper.

Engineering asset management (EAM) process is a broad discipline and the EAM functions and processes are characterized by its distributed nature. However, engineering asset nowadays mostly relies on self-maintained experiential rule-bases and periodic maintenance, which is lacking a collaborative engineering approach. To enrich the maintenance efficiency and customer relationship, this research proposes collaborative environment integrated by service center with good diagnosis and prognosis expertise. The collaborative maintenance chain jointly combines asset operation sites (i.e., maintenance demanders), service center (i.e., the system provider and maintenance coordinator), first tier collaborator (i.e., maintenance providers), and maintenance part suppliers. Meanwhile, to realize the automation of communication and negotiation among organizations, multi-agent system (MAS) technique is applied. With agent-based collaborative environment, the entire service level of engineering asset maintenance chain is increased. Moreover, during the MAS design processes, this research combines Prometheus MAS modeling methodology with Petri-net modeling methodology and unified modeling language (UML) to ease the design processes of MAS. The major contributions of this research contain developing a Petri-net enabled Prometheus multi-agent system (MAS) modeling methodology and construct an agent-based maintenance chain framework for integrated engineering asset management.

With the trend of globalization, enterprises require to extend their logistics scope from domestic to international, integrate logistics networks, and improve the agility and efficiency of global operations. This paper reviews the design and implementation of the industrial logistics hubs which belong to six manufacturing sectors in Taiwan over the past five years. Different industries have their own industry character, business strategy, and logistics models. Therefore, this paper organizes the executive problems and their corresponding solutions during project execution phases. Finally, this paper depicts the companies' executive experience and presents the future outlook for other companies and industry sectors' reference.

In hospitals and other medical facilities, nursing staff are responsible for the care of critically ill patients. Many hospitals use information technology to support nursing care. However, most medical information systems use manually recorded vital sign data for patient care and control. Medical staff cannot automatically detect abnormalities and provide consistent and immediate health care services. Thus, in this research, a Mobile Intelligent Medical System (MIMS) is developed that supports mobile nursing applications and clinical decision support. The functions include RFID-based mobile applications for monitoring physiological instruments. A Java-based expert system integrated with an RFID-enabled patient data collection module and a rule base is used to issue warnings and send diagnostic messages. The MIMS network model demonstrates how to bring more efficient services to patients while increasing safety and quality in a dispersed medical environment.

Reverse logistics research is used to analyze the processes associated with the flows of products, components and materials from end users to re-users in different industries. The products and components collected for reverse logistics are often widely dispersed, which complicates efforts to efficiently collect, reuse and reassemble used components for reprocessing and remanufacturing. Therefore, radio frequency identification (RFID) technology combined with the EPCglobal network architecture is applied to facilitate product and component data collection and data transmission. This research proposes a decision support model that integrates fuzzy cognitive maps trained using a genetic algorithm. The advantage of using fuzzy cognitive maps is that the model and the relationships among nodes (states) can be linguistically expressed both quantitatively and qualitatively. Furthermore, to diminish the subjective effects of the weights, the genetic algorithm is applied. The model and the information system integrate the EPCglobal network architecture with the RFID technology. Finally, a case concerning automobile repair reverse logistics is used to demonstrate the usefulness of the approach.

This correspondence presents a novel hierarchical clustering approach for knowledge document self-organization, particularly for patent analysis. Current keyword-based methodologies for document content management tend to be inconsistent and ineffective when partial meanings of the technical content are used for cluster analysis. Thus, a new methodology to automatically interpret and cluster knowledge documents using an ontology schema is presented. Moreover, a fuzzy logic control approach is used to match suitable document cluster(s) for given patents based on their derived ontological semantic webs. Finally, three case studies are used to test the approach. The first test case analyzed and clustered 100 patents for chemical and mechanical polishing retrieved from the World Intellectual Property Organization (WIPO). The second test case analyzed and clustered 100 patent news articles retrieved from online Web sites. The third case analyzed and clustered 100 patents for radio-frequency identification retrieved from WIPO. The results show that the fuzzy ontology-based document clustering approach outperforms the K-means approach in precision, recall, F-measure, and Shannon's entropy.

Owning to the rapid growth of consumer electronics market, there are urgent needs for better control over IC design projects, reuses of design knowledge (e.g., silicon intellectual property - SIP), and design collaboration in a virtual design platform. In order to provide IC industry a well-integrated design environment, this research proposes an information platform which combines the project management module, the design knowledge management module and the collaborative working environment for the IC industry. With the help of the project management module, the R&D team leaders can better control its partners' and its own schedules to shorten the time to market. Moreover, the design knowledge (SIP) management module provides the IC companies and their partners a knowledge exchange environment to enrich the design chain productivity and efficiency. Finally, the integration of collaborative working environment is the propeller to enable collaborative design. With the collaborative project and knowledge management platform, IC companies gain competitive advantages when working as a virtual design team.

In an era of rapid information expansion, people encounter huge amounts of intellectual property (IP) such as patents in digital format. These documents are usually too numerous to be fully utilized in R&D for new product designs. Therefore, efficient and effective ways of acquiring, organizing and presenting IPs (e.g., patent documents) have become very important for enterprises. In this paper, we propose a patent document summarization system using an integrated approach of key-phrase recognition and significant information density. Significant information density or information mass is calculated based on the summation of key-phrases, their relevant phrases, title phrases, domain-specific phrases, indicator phrases and topic sentences, divided by the total number of phrases in a paragraph or a document. External text mining game, compression ratio and retention ratio are used in system experiment and evaluation. This research enables enterprises to organize knowledge and intellectual assets efficiently and to peruse IP documents effectively.

In order to process huge amount of explicit knowledge documents in an organized manner, automatic document categorization is an important research area in the R&D knowledge management domain. In this paper, we propose a new document classification methodology based on neural network technology. We first extract key phrases from the document set by means of text processing and determine the significance of key phrases by their appearance frequency. After significant phrases are extracted, a keyword correlation analysis model is applied to compute similarity between key phrases. Then, synonyms are extracted from highly similar terms. The backpropagation neural network model is adopted as a classifier. The target output is to identify a document's proper category based on the hierarchical document classification scheme, i.e., the international patent classification (IPC) standard. In this research, patents related to designs of innovative power hand-tools are studied in their IPC classification scheme. Any related patent can be automatically and accurately classified using the pretrained neural network models. In the prototype system, we provide two modules for explicit knowledge management. The automatic classification module helps the user classify patent documents and the search module helps users find the correct patent documents quickly. The result shows a very significant improvement in document classification and identification in explicit knowledge management.