Thesis (M.Phil.)--City University of Hong Kong, 2005. / "Submitted to Department of Building and Construction in partial fulfillment of the requirements for the degree of Master of Philosophy" Includes bibliographical references (leaves 239-241)

No description available.

The second marketing model addresses interpretation issues of traditional multiple dimensional scaling (MDS) model, and has three key features. First, it can provide an explicit meaning to each dimension. Second, it can directly provide a link between brand positions and brand attributes, so that marketing mix can be designed to optimize share or profit. Third, for a case of n brands and m product attributes, we only need to estimate m nonzero parameters to represent all brand positions in the reduced space of d dimensions where d < m. The model is applied to a real data set of thirty four cars and nine attributes. Empirical findings suggest that the proposed model is a viable alternative to existing methods. / This thesis proposes two marketing models. The first model is a multiple criteria ranking (MCR) model. It proposes a consensus approach to solve multiple criteria ranking problems. Marketing decisions frequently involve ranking various subjects using multiple criteria. However, it is difficult to have acceptance from all subjects. The proposed multiple criteria ranking model allows each subject to have a "say" in the evaluation process. All subjects together resolve their conflicts through voting so that consensus ranking can be derived under the majority rule principle. The new ranking procedure can be formulated as a series of mathematical programming models, and illustrated with two cases. First, ranking the research performance of thirty one marketing scholars according to their publications in four leading marketing journals. Second, ranking the performance of one hundred information technology companies. / Lam Pong Yuen. / "June 2000." / Added t.p. in Chinese. / Director: Lau Kin Nam. / Source: Dissertation Abstracts International, Volume: 61-08, Section: A, page: 3259. / Thesis (Ph.D.)--Chinese University of Hong Kong, 2000. / Electronic reproduction. Hong Kong : Chinese University of Hong Kong, [2012] System requirements: Adobe Acrobat Reader. Available via World Wide Web. / Electronic reproduction. Ann Arbor, MI : ProQuest dissertations and theses, [200-] System requirements: Adobe Acrobat Reader. Available via World Wide Web. / Electronic reproduction. Ann Arbor, MI : ProQuest Information and Learning Company, [200-] System requirements: Adobe Acrobat Reader. Available via World Wide Web. / Abstracts in English and Chinese. / School code: 1307.

Quality Capability Self-diagnosis is a convenient and economical way to assess the performance of an operating quality system, as well as a basis for initiating necessary corrective actions essential to quality improvement and preparation for certification. This paper describes the research leading to the development of a cost-effective and systematic methodology for performing quality capability self-diagnosis. ISO 9000 series standards and the methods used to implement multicriteria system evaluation are employed to provide a sound basis for the development of this quality capability self-diagnosis scheme (QCSDS). The QCSDS has been developed to assist manufacturers in the conduct of quality assurance audits using internal personnel. The methodological structure of QCSDS is presented in two major parts: a regular model and a refined model. The regular model includes: (1) development of quality system auditing criteria, (2) selection of a suitable checklist developed from ISO 9000 series requirements, (3) development of importance weights for applicable criteria, (4) performance measurement, (5) quality system rating, (6) analysis of quality auditing results, and (7) suggestions for improvement. The refined model is developed to strengthen capability of the model and its reliability for confirming the effectiveness of an operating quality system using quality cost analysis, utility theory and regression analysis. A decision support system (QCSDDSS) based on the Quattro Pro spreadsheet is incorporated to facilitate the application of QCSDS. The QCSDDSS development is based on the regular model using ISO 9002 to provide both tabular and graphical displays for performance demonstration and improvement analysis. / Graduation date: 1994

In this study, an integrated solution methodology for a general discrete multi-criteria decision making problem is developed based on the well-known outranking method Promethee II. While the methodology handles the existence of interdependency between the criteria, it can also incorporate the prospect theory in order to correctly reflect the decision behavior of the decision maker. A software is also developed for the application of the methodology and some applications are performed and presented.

No description available.

This study is concerned with a sorting problem / the placement of alternatives into preference classes in the existence of multiple criteria. An interactive model is developed to address the problem, assuming that the decision maker has an underlying utility function which is linear. A recent methodology, Even-Swaps, which is based on value tradeoff is utilized in the model for both making an estimation of the underlying utility function and generating possible dominance among the alternatives on which it is performed. Convex combinations, dominance relations, weight space reduction, Even-Swaps and direct decision maker placements are utilized to place alternatives in preference classes. The proposed algorithm is experimented with randomly generated alternative sets having different characteristics.

Bi-level programming techniques are developed for decentralized decision problems with decision makers located in a two-level decision making system; the upper decision maker is termed the leader while the lower is the follower. Both the leader and the follower try to optimise their own objective functions and the corresponding decisions do not control but do affect those of the other level. This research aims at solving bi-level decision problems with five extensions, i.e. multiple leaders/followers/objectives, fuzzy coefficients and goals. By using particle swarm optimisation and/or cut set and/or goal programming and/or Nash equilibrium concept, related mathematical models and corresponding algorithms are developed to solve fuzzy linear bi-level decision problems, fuzzy linear multi-objective bi-level decision problems, fuzzy linear multi-follower multi-objective bi-level decision problems, fuzzy linear goal bi-level decision problems, multi-leader one-follower bi-level decision problems, one-leader multi-follower bi-level decision problems, and multileader multi-follower bi-level decision problems. A fuzzy bi-level decision support system is then developed which implements all the algorithms to support bi-level decision making with different features. Finally, by using these bi-level models and algorithms, we explore possible applications in the fields of railway train set organisation, railway wagon flow management, strategic bidding in the electricity market, and supply chains to solve real world bi-level decision problems. The results of experiments show that the models and algorithms are effective for solving real world bi-level decision problems.

Bi-level programming techniques are developed for decentralized decision problems with decision makers located in a two-level decision making system; the upper decision maker is termed the leader while the lower is the follower. Both the leader and the follower try to optimise their own objective functions and the corresponding decisions do not control but do affect those of the other level. This research aims at solving bi-level decision problems with five extensions, i.e. multiple leaders/followers/objectives, fuzzy coefficients and goals. By using particle swarm optimisation and/or cut set and/or goal programming and/or Nash equilibrium concept, related mathematical models and corresponding algorithms are developed to solve fuzzy linear bi-level decision problems, fuzzy linear multi-objective bi-level decision problems, fuzzy linear multi-follower multi-objective bi-level decision problems, fuzzy linear goal bi-level decision problems, multi-leader one-follower bi-level decision problems, one-leader multi-follower bi-level decision problems, and multileader multi-follower bi-level decision problems. A fuzzy bi-level decision support system is then developed which implements all the algorithms to support bi-level decision making with different features. Finally, by using these bi-level models and algorithms, we explore possible applications in the fields of railway train set organisation, railway wagon flow management, strategic bidding in the electricity market, and supply chains to solve real world bi-level decision problems. The results of experiments show that the models and algorithms are effective for solving real world bi-level decision problems.

Thesis (Ph. D.)--Victoria University (Melbourne, Vic.), 2008. / Includes bibliographical references.