Alfonso Bastias, Ph.D.
Profesor Facultad de Ingenieria
Universidad del DesarrolloEducación
Doctor of Philosophy (Ph.D.)
Dissertation's title:
"Towards the Application of Learning Systems for Decision Support in Construction Engineering and Management".
Committee:
- Dr. Luis F. Alarcon. Profesor Titular. School of Engineering. Pontifical Catholic University of Chile.
- Dr. David Ashley. President of University of Nevada, Las Vegas.
- Dr. James Diekmann. K. Stanton Lewis Professor of Construction Engineering and Management. University of Colorado.
- Dr. Keith Molenaar (Chair - Advisor). Assistant professor at University of Colorado.
- Dr. Kenneth Strzepek. Associate professor. University of Colorado
Proposal's abstract
Construction managers need to make decisions in an uncertain environment, where unexpected variables are present everyday and everywhere. To adequately support their decisions and decrease any negative impact and collateral effect, they use computational tools called decision support systems (DSS). DSS are widely used in the construction industry, presenting many advantages such as historical data available with fast processing of information, accuracy and effectiveness of output with a friendly interface.
However, a review of ninety-three DSS in the construction over the past 30 years showed that most of them are static, where the model has fixed its parameters, and member functions. Static models can quickly become obsolete; requiring manual adjustment to be relevant in a dynamic environment such as the construction engineering and management field. A better approach to solving the problem of changes of the decision environment within the construction industry is to develop dynamic models based on learning systems.
This research explores the application of learning capabilities in decision support systems in the construction industry by examining questions such as: what is the history and current state of DSS in construction engineering and management research?, what are the key components of a learning system for decision support?, and what are the characteristics of data in the construction engineering and management industry that must be addressed to create a general framework for applying learning systems? The outcome of this research is a general framework to apply the learning component into decision support systems for the construction engineering and management field.
Final Defense: June, 28th 2006
Magíster en Ciencias de la Ingeniería (M.Sc.)
thesis's title:
"Un Ambiente Integrado para la Modelacion de Decisiones Estratégicas".
Committee:
- Dr. Luis F. Alarcon (Chair - Advisor). Profesor Titular. School of Engineering. Pontifical Catholic University of Chile.
- Dr. David Fuller. Profesor Adjunto. Pontifical Catholic University of Chile
- Dr. Raul O'Ryan. Profesor Asociado. School of Engineering. University of Chile.
- Dr. Luis Contesse. Profesor Titular. Pontifical Catholic University of Chile
Abstract
A computer environment to model and evaluate strategic decisions is presented. The system implements modeling concepts originally developed to evaluate project execution strategies, extending and generalizing the modeling methodology to a broader range of strategic decisions.
The computer system is designed to help the users in building a conceptual model for the decision problem, this model is a simplified structure of the variables and interactions that influence the decisions being analyzed, including internal as well as external factors. The mathematical component uses concepts of cross-impact analysis and probabilistic inference to capture uncertainties and interactions among project variables. The system provides multiple analysis capabilities, including sensitivity analysis, selected outcome prediction, isolated or combined effect of strategies, and changes in performance due to changes in the external environment. The models allow management to test different decisions or strategies and predict expected outcomes with a reduced modeling effort.
Defensa Final: 30 de September 1998 (con Distinción Máxima)