Wyniki wyszukiwania - BazTech

1

Automatic Enhancement of the Reference Set for Multi-Criteria Sorting in The Frame of Theseus Method

Fernandez E., Navarro J., Salomon E.

Foundations of Computing and Decision Sciences

|

2014

|

Vol. 39, No. 2

57--77

EN

Some recent works have established the importance of handling abundant reference information in multi-criteria sorting problems. More valid information allows a better characterization of the agent’s assignment policy, which can lead to an improved decision support. However, sometimes information for enhancing the reference set may be not available, or may be too expensive. This paper explores an automatic mode of enhancing the reference set in the framework of the THESEUS multi-criteria sorting method. Some performance measures are defined in order to test results of the enhancement. Several theoretical arguments and practical experiments are provided here, supporting a basic advantage of the automatic enhancement: a reduction of the vagueness measure that improves the THESEUS accuracy, without additional efforts from the decision agent. The experiments suggest that the errors coming from inadequate automatic assignments can be kept at a manageable level.

2

Evolutionary multi-objective optimization for inferring outranking model’s parameters under scarce reference information and effects of reinforced preference

Fernandez E., Navarro J., Mazcorro G.

Foundations of Computing and Decision Sciences

|

2012

|

Vol. 37, No. 3

163--184

EN

Methods based on fuzzy outranking relations constitute one of the main approaches to multiple criteria decision problems. The use of ELECTRE methods require the elicitation of a large number of parameters (weights and different thresholds) but direct eliciting is often a demanding task for the decision-maker (DM). For handling intensity-of-preference effects on concordance levels, a generalized concordance model was proposed by Roy and Slowinski which is more complex than previous outranking models. In this paper, an evolutionary multi-objective-based indirect elicitation of the complete ELECTRE III model-parameter set is proposed. The evolutionary multi-objective inference method is successfully extended to inferring reinforced-preference model parameters. Wide experimental evidence is provided to support the proposal, which performs well even working on small size reference sets.

3

Bismuth Hall thruster I: Simulation-suggested design and performance

Scharfe D. B., Scharfe M. K., Capelli M. A., Fernandez E.

Journal of Technical Physics

|

2008

|

Vol. 49, no 3-4

329-348

EN

A two-dimensional radial-axial hybrid simulation of the xenon-fueled Stanford Hall Thruster has been adapted to model a bismuth-fed thruster with varying channel geometry. The simulation treats the electrons as a quasi-one-dimensional fluid and the neutrals and ions as discrete superparticles advanced using a particle-in-cell (PIC) approach. Since experimental data of the electron cross-field mobility does not exist for the bismuth-fueled thruster, a model for electron transport based on shear suppression of plasma turbulence is used to compute a mobility from simulated plasma properties. While the bismuth propellant showed poor performance with an 8 cm channel length, results improved significantly as the simulated channel was shortened to 3.3 and 2.4 cm. The simulation of bismuth propellant at the shortest channel length provided significantly improved ionization fraction, thrust, efficiency, and thrust-to-power compared to xenon propellant on either the 8 cm or 2.4 cm channel, as can be expected due to the higher atomic mass and lower ionization potential of bismuth. With results indicating that optimal performance of the bismuth thruster occurs with a sub-3 cm channel length, such a design is suggested for a developing laboratory-model bismuth thruster.

4

Computer-based decision models for R&D project selection in public organizations

Fernandez E., Navarro J.

Foundations of Computing and Decision Sciences

|

2005

|

Vol. 30, No. 2

103-131

EN

Project selection is the most important problem concerning R&D management in public organizations, where weak heuristics are used for evaluating projects and making decisions about final portfolios. We propose here an integrated approach for analyzing projects and solving portfolio problems whose central parts are the use of decision tables as models of decision-maker's preferences and beliefs, and a mode! of R&D portfolio quality derived from Utility Theory and based on fuzzy sets to model some sources of imprecision. The resulting optimization problem is very complex in order to be solved by classical mathematical programming methods, so we propose an evolutionary algorithm able to achieve a strong improvement of the quality of solution. Some results are applicable in other problems outside the scope of this paper.

5

An improved method for deriving final ranking from a fuzzy preference relation via multiobjective optimization

Fernandez E., Olmedo R.

Foundations of Computing and Decision Sciences

|

2003

|

Vol. 28, No. 3

143-157

EN

Methods for deriving final ranking from a fuzzy preference relation do not perform well in presence of irrelevant alternatives or in case of complex graphs with numerous circuits. Recently some approaches based on the idea of reducing differences between a global model of preferences and a final ranking via multiobjective optimization with an evolutionary algorithm have been proposed. In this work a new method is presented based on similar ideas but improving them. The multiobjective optimization problem is separated into two steps and solved with a better model of preferences, also using an evolutionary algorithm simpler than the former. These improvements allow us to obtain better compromise solutions in a simpler way than the previous proposals.

6

A normative proposal for obtaining a suitable multiattribute generalized cost in routing problems

Lopez F., Fernandez E.

Foundations of Computing and Decision Sciences

|

2002

|

Vol. 27, No. 1

17-26

EN

Up to now, the routing problem with physical planning, (choosing the best route linking each pair of adjacent nodes belonging to a road network), has been solved only using an imprecise heuristic approach. In this paper we present a normative method which uses some proxy variables for modeling the main attributes, and is based on a set of rational axioms which provides a good framework for clearing Decision Maker's preferences and beliefs, following the paradigm of Decision Analysis. This method has been successfully applied in practical decision making.