Wyniki wyszukiwania - BazTech

Ograniczanie wyników

Znaleziono wyników: 2

Liczba wyników na stronie

Wyniki wyszukiwania

Wyszukiwano:
w słowach kluczowych: logic-algebraic method

Sortuj według:

Ogranicz wyniki do:

CP-driven approach to multicriteria decision making based on imprecise data

Bach I., Bocewicz G.

Foundations of Computing and Decision Sciences

2008

Vol. 33, No. 2

119-130

Multi-criteria decision-making encompass different aspects of Small and Medium Size Enterprise functioning, e.g. money flow, personnel allocation, task oriented scheduling, etc. In general multi-criteria decision making is aimed at determining an company preferences (e.g. financial benefits, intangible benefits, availability of resources, risk level, etc.), and customers requirements (e.g. specified by the time of implementation and budget of the relevant production orders). Moreover all decisions are usually time and financially constrained. Considering above, decision problems are specified by the diverse character of information (including distinct and imprecise data). The approach considered regards of the Logic-Algebraic Method based and Constraint Programming methodology aimed at interactive and multi-criteria decision making. In case the data introduced are specified by a membership function its representation can be discretized and replaced by an ordered set of discrete values. It means, instead of standard fuzzy-set-like operations, e.g. fuzzy complement, intersection, union, and fuzzy inference rules, a set of constraints is considered. In such approach both: distinct and imprecise data as well as linking them relations are treated in a unified form of discrete Constrained Satisfaction Problem. Moreover, implementation of multi-criteria decision making directly follows from the nature of constraint programming paradigm (constraints propagation and variables distribution). The way of possible approach implementation is illustrated in the example enclosed.

Execution of parallel logic-algebraic algorithms in expert system SymLog

Borzemski L., Fraś M.

Journal of Applied Computer Science

2002

Vol. 10, nr 1

7-22

This paper presents the SymLog system, which is at the same time a full featured expert system for execution of expert tasks formulated following the rules of the logic-algebraic method, as well as the simulation tool for studying parallel execution of expert tasks in multiprocessor environments.