Flexible programming : modelling uncertainty and conflict

By: Benseman, B.R. (DSIR Physical Sciences, Applied Mathematics Group. Wellington).
Contributor(s): DSIR Physical Sciences, Applied Mathematics Group. Wellington.
Material type: materialTypeLabelBookSeries: DSIR Physical Sciences report ; 55.Publisher: Wellington : DSIR Physical Sciences, Applied Mathematics Group, 1992Description: 20 p.ISSN: 1170-4438.Subject(s): FUZZY SETS | OPTIMIZATION | COMPUTER SIMULATION | MATHEMATICAL MODELS | LINEAR PROGRAMMING | STANDARDS
Incomplete contents:
This paper out lines the use of fuzzy mathematics in the optimisation area, and shows how combining membership functions from various fuzzy constraints is confusing.However the key is that goals and constraints are similar and are not often sharply defined. Flexible LP lets us handle fuzzy and conflicting objectives, and use soft constraints for uncertain constraints. Flexible LP saves us time because we avoid the problem of finding the weights of conflicting goals, to combine them into one common measure. Finally we don't need to treble the number of decision variables like stochastic programming does to model different probability levels. We can model these conflicting objectives and uncertain or secondary constraints with just one extra variable. (auth.)
Tags from this library: No tags from this library for this title. Log in to add tags.
Item type Current location Call number Copy number Status Date due Barcode
ABSTRACT ABSTRACT NIWA BIBLIOGRAPHY
1 Available 88811-1001

1 ref; 6 figs

This paper out lines the use of fuzzy mathematics in the optimisation area, and shows how combining membership functions from various fuzzy constraints is confusing.However the key is that goals and constraints are similar and are not often sharply defined. Flexible LP lets us handle fuzzy and conflicting objectives, and use soft constraints for uncertain constraints. Flexible LP saves us time because we avoid the problem of finding the weights of conflicting goals, to combine them into one common measure. Finally we don't need to treble the number of decision variables like stochastic programming does to model different probability levels. We can model these conflicting objectives and uncertain or secondary constraints with just one extra variable. (auth.)

AM

PY

There are no comments on this title.

to post a comment.

Powered by Koha