CityU Institutional Repository >
CityU Electronic Theses and Dissertations >
ETD - Dept. of Information Systems >
IS - Doctor of Philosophy >
Please use this identifier to cite or link to this item:
|Title: ||Decision support methods for R&D project selection|
|Other Titles: ||Ke yan xiang mu xuan ze de jue ce zhi chi fang fa|
|Authors: ||Sun, Yonghong (孫永洪)|
|Department: ||Department of Information Systems|
|Degree: ||Doctor of Philosophy|
|Issue Date: ||2009|
|Publisher: ||City University of Hong Kong|
|Subjects: ||Decision support systems.|
Research and development projects.
Research -- Decision making.
|Notes: ||CityU Call Number: T58.62 .S86 2009|
vi, 113 leaves : ill. 30 cm.
Thesis (Ph.D.)--City University of Hong Kong, 2009.
Includes bibliographical references (leaves 94-107)
|Abstract: ||This dissertation proposes decision support methods to facilitate the decision tasks in research and development (R&D) project selection. Project selection for funding is an important task for government and private funding agencies. Usually, external experts are invited to review submitted proposals, and then funding decisions are made based mainly on their opinions. Finding a good match between proposals and experts is the key issue of project selection since it requires knowledge of experts’ research areas and their expertise levels. It is also necessary to preprocess submitted proposals in advance in order to complete the assignment of proposals to experts. This dissertation addresses three problems: expert evaluation, proposal grouping, and assignment of proposals to experts.
First, this study presents a group decision support approach to evaluate experts. A formal procedure that integrates both objective and subjective information on experts is also presented. A group decision support system is designed and implemented for illustration of the proposed method. Second, to deal with the huge number of proposals, it is common to group them according to their similarities in research areas. This study presents a novel ontology-based text mining approach to clustering research proposals. Third, a key step is to assign the most appropriate experts to proposals for which a hybrid knowledge and model approach is proposed.
Evaluation procedures are implemented for the three proposed approaches. First, for expert evaluation, a questionnaire survey is conducted. User perception on the proposed approach and system is investigated, for which the statistical results indicate that user comments are positive. Second, for proposal grouping, historical data in the National Natural Science Foundation of China (NSFC) is collected as input to implement the proposed approach. The test results show that the average similarity degree of the proposals in each group is quite high. Third, for the assignment problem, simulated data is used to test the proposed method. The results show that the match degree between research proposals and reviewers is very high. Also, the workload of each reviewer is well balanced. Thus our validations are supported by both theoretical analysis and practical experiments with real data.|
|Online Catalog Link: ||http://lib.cityu.edu.hk/record=b3947579|
|Appears in Collections:||IS - Doctor of Philosophy |
Items in CityU IR are protected by copyright, with all rights reserved, unless otherwise indicated.