ARCOM

Search or browse again.

Click on the titles below to expand the information about each abstract.
Viewing 6 results ...

Inyang, E D (1983) Some aspects of risk analysis for decision making in engineering projects management, Unpublished PhD Thesis, , The University of Manchester (United Kingdom).

Luo, J (1998) The transitional construction industry and sino-foreign construction joint ventures in the p.R. Of China, Unpublished PhD Thesis, , The University of Manchester (United Kingdom).

Ng, S-t T (1996) Case-based reasoning decision support for contractor prequalification, Unpublished PhD Thesis, , The University of Manchester (United Kingdom).

Nosair, I A R (1987) The relevance of computerised modelling techniques to construction management problems and training in Egypt, Unpublished PhD Thesis, , The University of Manchester (United Kingdom).

Parvar, J (2003) Neural networks decision support system (decision to bid), Unpublished PhD Thesis, , The University of Manchester (United Kingdom).

• Type: Thesis
• Keywords: accuracy; decision support; failure; rationality; automation; communication; decision making; probability; professional; neural network; regression model
• ISBN/ISSN:
• URL: https://www.proquest.com/docview/1820624652
• Abstract:
  Initially, the research focused on the identification of those factors that affect valid decision-making in the decision domain (the decision to bid process). A critical review and analyses of the literature, and further consultation with professionals in the decision domain assisted in developing a process hierarchy diagram which identifies 21 factors as important for decision making in the domain. The process hierarchy diagram develops a conceptual view of the important factors and their relationships, which can be shared and related to by target users of the research.A questionnaire and data collection form (pro-forma) were developed, tested and enhanced to ensure effective communication with the target audience. Using the developed questionnaire, data were collected from historical projects of a contracting company.Regression models were used to model the collected data set. Regression models failed to provide an acceptable degree of accuracy for modelling the collected data set. The failure of regression models is an indicator that complex and non-linear relationships are involved in the decision domain, which their linear approximation cannot provide an acceptable accuracy for modelling and prediction.Neural networks are employed to model the collected dataset. Neural networks model the decision domain with a high accuracy of prediction. The developed neural network is further enhanced, through adding a user-friendly user interface, to function as a DSS for the decision to bid. The developed DSS can function as a demonstrator to communicate the potentials of neural networks for automation of semi-structure knowledge domains.The Decision Support System (DSS) for the decision to bid process was developed from the data collected from the contracting company, and was supported by a neural networks system that emulates the decision making approach by that specific contracting company. It is possible to argue that the approach to decision making for the decision to bid process by one contracting company might not be the best or optimum practice. Therefore, bringing the validity, both rationality and optimality, of the DSS recommendations under question, which can damage its usability and acceptability.To ensure that the DSS recommendations have a wide spread acceptability, a rational and optimal model of decision making for the decision to bid, based on the important factor set identified, needed to be developed. The rationales for the model must be so evident and strong that makes it readily acceptable by the practitioners in the field.A rational and optimal model of decision-making for the decision to bid process was developed. The rationales for the model are very evident, simple and perceived to have wide acceptability. The model is devised to improve the effectiveness of the decision-making process for the decision to bid process, through focusing resources to the projects which have a high probability of success, and are most suitable to achieve the strategic objectives of the organization. Professionals in the field assessed the rational and optimal model of decision making for the decision to bid as valid and an effective approach to decision making.A neural networks system with high accuracy of prediction is developed to automate the rational and optimal model of decision making for the decision to bid process. The developed neural networks system is further integrated with a user interface to function as a DSS.The developed rational and optimal model of the decision-making and the DSS have been assessed and evaluated by professionals in the field and held to be a valid and an effective approach to decision making. Also, the DSS is being used to support decision making for the decision to bid process in their organizations.This research presents and recommends the adoption of an effective system development methodology for the development of neural networks decision support systems. The recommended methodology is the outcome of successfully applying neural networks to the automation of the decision to bid rocess, and from the development of a neural networks Decision Support System (DSS) to support effective decision making for the decision to bid.

Samo, S R (1999) Energy conservation in UK housing and the effect of building regulations, Unpublished PhD Thesis, , The University of Manchester (United Kingdom).