Challenges in Information Technology Management

PREFACE

CONTENTS

In recognition of the increasing importance of rapidly changing needs and opportunities, today's global markets require interoperability to integrate diverse logistics processes to share information and collaborate among logistics partners. The multi-agents system provides a promising computing paradigm for efficient logistics service response and integration of logistics service processes. This paper discusses an approach to designing an integrated logistic system based on multi-agents. A framework of an integrated logistics process is proposed. A prototype based on the proposed framework is implemented using ZEUS toolkit. It presents how the proposed integrated architecture can be used to establish collaborative logistics business environment by providing dynamic processes services.

Mining association rules are used to analyze the data in a database to discover interesting rules. The algorithms for mining association rules with composite items have the potential to discover rules which cannot be found out by algorithms without composite items. Algorithms for finding large composite items should scan the database for every candidate composite item to determine whether it is large. In this paper, we design some improved algorithms for finding large composite items which only need to scan the database one time to find all the large composite items. This algorithm also allows the reduction of many more redundant candidate composite items.

This paper introduces an ontology mapping approach based on triangle fuzzy expression. This approach imports several kinds of similarity fuzzy expressions based on concept feature information. These similarity fuzzy expressions can be used to extend traditional mapping based on name and instance to composite mapping based on name, instance, attribute, taxonomy, dependence and axiom etc. To implement this approach, two questions should be answered: the first is the unification of all kinds of similarity fuzzy expressions and the second is how to realize the composite mapping based on the unified similarity fuzzy expression. The triangle fuzzy expression is regarded as the criterion of the other similarity fuzzy expressions. After the transformation from the other similarity fuzzy expressions to the triangle fuzzy expression, the unified similarity fuzzy expression is realized. Simultaneously, a mapping computation tree is also introduced to finish the whole composite mapping through a series of weight configuration, unification and synthesis. Finally, associated with traditional non-fuzzy mapping approach, the mapping accuracy of this approach will be evaluated through an application example.

Decision tables are a simple yet important and powerful tool to help implement the complex logic of many domains. However, the definition of concepts, variables and rules that underlie a decision table often are or become implicit. As the decision tables grow, ambiguities, inconsistencies and conceptual reasoning difficulties arise and building them becomes a time-consuming and difficult task. In order to address the problems, we use ontologies to represent, store and maintain the domain concepts. A decision table annotated with domain ontologies is called Semantic Decision Table (SDT). The rules that are generated from the formal commitment, together with the existing domain ontologies constitute the knowledge kernel of SDT. SDT is modeled in the DOGMA framework. It contains plausible binary facts called SDT lexons, and the SDT commitments, which the agents use to constrain and use the lexons. To facilitate reasoning, we encode the SDT commitments in Prolog prototype. In this paper, we focus on how to write SDT commitments in Prolog. Furthermore, we illustrate the design of the tool and its implementation.

Genetic algorithms are promising search techniques dealing with the combinatorial optimization problems. The main disadvantage of GA is the slow convergence rate of the search. In order to improve convergence rate of the search, the former researchers put forward two common strategies, one is local search methods and the other is parallelization. This paper tries to combine the two strategies and design the architecture of parallel hybrid GA. We propose three parallel hybrid GAS: (1) GA holds the shared population, (2) local search holds the shared population, and (3) independent shared population. This paper selects message passing interface (MPI) to implement the parallel hybrid GAS program. We compare our parallel hybrid GA with the pure GA and the pipelining hybrid GA, and the results of the simulation show that parallel hybrid GAS can combine the merits of the local search approach and the parallel mechanism.

The purpose of this study is to introduce the innovated technology of ultrasonic cleaning equipment with the mechanism of the cooperation between the academic circles and businesses. The study focuses on the identification of the local businesses with the equipment and a survey was made to investigate how the equipment is accepted in the market. We target local hi-tech businesses having clean rooms (class ≦ 10000) and extract the constructs influencing the acceptance in the market by Delphi Method and through Analytic Hierarchy Process (AHP), we calculate the relative weighting of each construct and principles to explore the future development of ultrasonic cleaning equipment. From our research, we found that two elements—government policies and publicity are the keys for local hi-tech businesses in selecting the ultrasonic cleaning equipment. Therefore it is suggested that manufacturers, when promoting the ultrasonic cleaning equipment, combine with the government publicity and regulations while enhancing the stability and quality of the equipment so that consumers can identify with the cleaning level after using the equipment. In the meantime, the study also predicts the future requirements and trend of the ultrasonic cleaning equipment and we believe there will be great market requirements with business opportunities.

In a supply chain environment, determining batch size at each stage is of importance. Using large batch size can save costs but time is wasted as trade-off. Small batch size can speed up the process but costs are increasing. To represent both aspects, this paper proposes the nonlinear integer mathematical model with the objective function of minimizing both costs and time simultaneously. This model can be used to determine (production) batch size to match up the flow of production at manufacturer with the demand at retailers and means of shipment (transfer batch size). Since the nature of this problem is NP-hard, at each stage of supply chain, the acceptable production and transfer batch sizes are obtained using the advantage of genetic algorithm approach. The numerical illustration and results show that it is not necessary to use the same batch size when manufacturing and transferring.

Customer relationship describes a sort of interactions between enterprises and their customers, and customer relationship management always emphasizes on the research of customer value: from enterprise to customer and from customer to enterprise. In this paper, cognition process is introduced to demonstrate the two types of values are bidirectional and to analyze the bidirectional values, and the interaction mechanism of bidirectional values is explained in terms of the associated processes. Moreover, the factors in bidirectional values are identified via analyzing the Customer Satisfaction Index Model, and concretize the interactive relationships of bidirectional values. Meanwhile, the mathematic model is set up to demonstrate the effect of interactive relationships. Thereby a sound and stable relationship between enterprises and their customers can be realized via combination of bidirectional values based on cognition process.

Mining sequential patterns for discovering frequent sequences has been widely studied as a data mining problem. A challenging research is to extend its use to data streams. A data steam is an unbounded, continuously generated sequence of data transactions. In this paper, we propose an online single-pass algorithm called OFSD (Online Frequent Sequence Discovery), to mine the set of all frequent sequences in a data stream whose frequency rates satisfy a minimum user defined frequency rate (f_u). The algorithm significantly reduces the number of elements in the candidate set (a set of candidate sequences that should be kept for further exploration) that efficiently increases its performance in comparison with other general solutions. The simulation results show the effects of f_u variation and the application defined threshold (C_M) on the frequent phrase detection process.

The impact of information technologies on negotiations is not limited to the use of electronic communication such as email, fax, telephone, etc. Information technology has changed the way a negotiation problem can be represented and a negotiation process structured. For example, e-commerce systems have created many more activities undertaken in negotiations, including, efficient matching of potential negotiators; exchange, comparison and categorization of information, and the use of tools for information gathering, problem structuring and analysis, and interpretation of offers. These new possibilities have led to the emergence of formal negotiation procedures and protocols, which are necessary for the use of rich and expressive information technologies in various stages of negotiation processes rather than solely for the exchange of messages. In this paper, we proposed a formal negotiation model for automated agents.

Constraint factors exist in any system will influence the whole value of the system. Bottleneck dilatation and bypass methods can only control Supply Chain (SC) bottlenecks. Therefore these two methods cannot solve the problem radically. As transformation to the traditional structure of SC, Supply and Demand Network with multi-functional and opening characteristics for enterprise (SDN) is a perfect method which can reduce the rate of SC bottleneck question from origin.

Requirement analysis is an important stages in information systems development (ISD) but is characterized by informality and uncertainty. In this paper, we will look at how well-known information systems methodologies handle it, and identify main problems and their causes. Then we argue that an explicit information analysis stage would be required in order to alleviate these problems, the information that an agent in a domain requires is identified and formulated in this stage. Such a stage would need a mechanism. We describe how hermeneutics might enable us to construct the mechanism whereby information is created and information flow takes place.

A recent surge of interest in social tagging, also known as folksonomy, challenges the formal and structured knowledge representation in the Semantic Web. Social tagging is attractive in its low barrier to entry and personal and community aspects. However, the benefits of social tagging come at the cost of reduced machine-understandable and reduce effectiveness in information retrieval and organization. To address the above limitations, we propose a framework, UNITAG, to enhance existing social tagging systems with semantic information. It is shown that the framework facilitates information sharing in folksonomy while retaining the usability of folksonomy.

The most important productivity in information age may be the knowledge and human resource. A growing number of people and enterprises pay attention to the knowledge management in organizations. We believe the IT assistant Knowledge management (KM) system serves as both an extension and an integration of traditional e-business, and thus, it is necessary for a firm's innovation and the maintenance of competitive power. However, many problems present in the way of the design and implementation of KM systems. By treating the KM system as a complex system, we find a new way to solve these problems. We design the architecture of KM system with the methodology of complex adaptive system, and resolve many practical difficulties to some extent. Furthermore, the problem in technology is merely one aspect of the whole thing, more works in management and organization, such as corresponding regulations on evaluation and promotion, are needed to make sure a smooth implement of a KM system.

Even though Enterprise Resource Planning (ERP) systems are increasingly implemented in today's business environment, results of ERP investments remain debatable. This study would like to investigate the financial impacts of ERP systems for the manufacturing sector in Taiwan. Findings demonstrate that most of performance indexes show no significant differences after ERP installation. Only two indexes such as “Inventory Turnover” and “Average Number Days Inventory in Stock” show significant improvement after 2 or 3 year ERP implementation. Other indexes such as “Return on Equity” and “Return on Assets” even have worse performance than before. This outcome implies that ERP systems have not shown significant benefits with respect to financial performance under our short-term observations. It may take longer period to see real financial advantages like the ERP vendors claimed.

Recent advances in technology have led to the wide availability of low cost wireless networking equipment. An increasing number of households are acquiring wireless networks. The low cost and convenience of not having wires around the house is appealing to users who want a tidy and clutter-free home environment. Simultaneously, there has been a growing attention towards security in wireless networks – both home networks and public networks. This study sets out to fill this gap through a quantitative and qualitative investigation. The results suggests that wireless equipment manufacturers can contribute to overall wireless security by providing simple security oriented user interfaces where options that violate best practices are hidden from the users.

Two important classes of ERP systems, real ERP systems and ERP simulation systems can be used to support ERP teaching and learning. While both systems help learners to have an understanding and experience of ERP, the features and applications of real ERP systems and ERP simulation systems are different. In this paper, we present a comprehensive comparison between these two classes of systems. We discuss the differences and similarities of SAP Business One, which is a Business ERP system, and SIMERP, which is an ERP Simulation system designed and implemented by Economic Science Laboratory, Renmin University of China. Both systems have been used to support ERP teaching and learning in Renmin University of China since 2004. We describe how the two types of systems can be used differently in education.

This paper briefly introduces the theory about supply chain, inventory management, and Jointly Managed Inventory (JMI). By a real-world case study, it discusses the status analysis, implement strategy and result evaluate of Jointly Managed Inventory (JMI) model about cigarette assistant material management in Shanghai Tobacco CORP.

Third party trust mechanisms are of great importance in establishing a secure and credible C2C transaction platform. Under C2C environment, these trust mechanisms include feedback system, an escrow service, C2C intermediary and ecommerce laws and regulations. This paper used a questionnaire survey, collecting 115 samples and analyzing it with PLS Graph3.0. Results showed that escrow services, C2C intermediary and feedback systems played significant roles in establishing consumer's trust in sellers, which in turn reduced consumer risk perception and enhanced his transaction intention.

Reputation can impel and restrict players' behavior, based on which, the paper regards reputation as the information reflecting the historic records and characteristics of players, and describe the dynamic mechanism to cooperate in the supply chain between the upstream and the downstream enterprises. Pursuant to the mechanism herein, agents' reputation are updated in accordance with Bayes' Law, therefore, the paper further establishes task distribution model and cooperation return rate model, through which, such problems as adverse selection and moral hazard can be effectively overcome before or after signing agreement.

Current Internet search engines do not always return results that users find useful, not least because the markup languages that are typically used only allow keyword searches. Semantic Web is a collection of web technologies that include a number of markup languages, such as RDF, OWL and RDFS, which can be used to model domain ontologies. Domain ontologies provide humans and computers (software agents) with a common understanding of a resource structure and help the creation of more effective web search systems. In this paper, we use Web Ontology Language (OWL) to model a travel domain ontology. Rather than inviting an expert to model the ontology, we create the travel ontology by collecting and analyzing the structural information from a number of travel related websites. The model is used to implement a tourist information portal, SemTour:HK, which is constructed using Semantic Web technologies.

This paper highlights the problems of information explosion and the incapability of currently available search engines in finding what we mostly want. In particularly, these search engines cannot offer users the facility of specifying the categories and time frames they receive and cannot provide the online news information with the required frequency. To address these problems, we present the design and implementation of- “Ai-Times”, a distributed web news retrieval system which can accurately retrieve and organize the web news information. We describe the optimized crawler algorithm, the news extraction algorithm, and explain how MapReduce is used in “Ai-Times” and can be improved to get better performance.

We describe a middleware framework for context-aware applications. The key component in our framework is the Service Adaptation Model. The middleware infers the best service policy by making use of the fuzzification of different kinds of contextual information and by measuring and comparing the degree of fit between the current context and a predefined optimal context. Our system enables automatic service adaptation in changing contexts and provides best QoS services for pervasive applications.

This paper proposes a project management model for customer relationship management projects for the service industry. It presents key findings from a survey of companies from the service industry and suggests that good communication management can increase the success rate of CRM projects. The other important project management areas are scope management, time management and project planning. The three key success factors are user involvement, executive commitment and process analysis. And the top three best practices for CRM project management are getting staff and management support and commitment, focusing on business not the technology, and managing stakeholder expectation. The findings confirm that people are critical to the success of CRM project management.

Modem software for Enterprise Application Integration (EAI) provides tools for modeling integration scenarios. A drawback of these tools is the lack of functionality for exchanging or integrating models from different EAI products. Consequently, developers are only partially able to describe real heterogeneous IT environments. Our goal is to avoid the creation of these so-called ‘integration islands’. For that purpose we present an approach which introduces an abstract view by technology-independent and multivendor-capable modeling for both development and maintenance. With this approach, we propose a toolset- and repository-based refinement of the abstract view to automate implementation with real products and deployment on real platforms.

Hong Kong CyberU (HKCyberU) is a company co-founded by the Hong Kong Polytechnic University (HKPolyU) and the Pacific Century CyberWorks (PCCW). As universities are investing in e-learning for Asynchronized Learning Networks (ALN), HKCyberU provides a business model for developing ALN in the Asia Pacific Rim. HKPolyU provides expertise and academic awards in professional education; PCCW provides investment, technical infrastructure, support, broadband capabilities, marketing and administration of courses. In addition to offering various HKPolyU awards, HKCyberU has developed strategic alliance with a number of universities in overseas and China to offer asynchronized/web supporting learning network. As higher education institutions are planning to finance e-Learning courses, HKCyberU model provides a good example of developing and offering courses online in Hong Kong and China.

Agent-oriented conceptual modeling notations such as i* represents an interesting approach for modeling early phase requirements which includes organizational contexts, stakeholder intentions and rationale. On the other hand, Unified Modeling Language (UML) is suitable for later phases of requirement capture which usually focus on completeness, consistency, and automated verification of functional requirements for the new system. In this paper, we propose a methodology to facilitate and support the combined use of notation for modeling requirement engineering process in a synergistic fashion. For organizational modeling/early phase requirements capturing we use the i* modeling framework that describes the organizational relationships among various actors and their rationales. For late (functional) requirements specification, we rely on UML Activity Diagram.

This paper highlights the situations that in recent years Chinese Higher Education Institutes (HEIS) have sought to improve their quality of education by attracting highly qualified faculty, training younger instructors, and using the new media and the Internet to deliver content. There have been few studies of the progress of these efforts, especially where they combine online and traditional teaching. This study describes students' responses to the introduction and use of an e-Learning platform or Course Management System (CMS) in an IT course in a Chinese HEI. The results suggest that e-Learning had no effect on the computer literacy of the participant students and that students who liked the blended instruction or Student Centered Teaching supported with e-Learning (SCeT) spent more time on their work and improved their IT skills.

Information systems Information systems (IS) are widely used but are rarely objectively evaluated. Further, there are difficulties associated with choosing the indices upon which an evaluation of IS should be made and with the numerical representation of human knowledge. This paper propose a novel IS evaluation system that uses Fuzzy Neural Networks (FNN) and provides a comprehensive assessment that draws data from every layer of the IS.

A key problem in high dimensional anomaly detection is that the time spent in constructing detectors by the means of generate-and-test is intolerable. In fact, due to the high sparsity of the data, it is ineffective to construct detectors in the whole data space. Previous investigations have shown that most essential patterns can be discovered in different subspaces. This inspires us to construct detectors in significant subspaces only for anomaly detection. We first use ENCLUS-based method to discover all significant subspaces and then use a greedy-growth algorithm to construct detectors in each subspace. The elements used to constitute a detector are grids instead of data points, which makes the time-consumption irrelevant to the size of the normal data. We test the effectiveness and efficiency of our method on both synthetic and benchmark datasets. The results reveal that our method is particularly useful in anomaly detection in high dimensional data spaces.

AUTHOR INDEX