Preferencje help
Widoczny [Schowaj] Abstrakt
Liczba wyników

Znaleziono wyników: 11

Liczba wyników na stronie
first rewind previous Strona / 1 next fast forward last
Wyniki wyszukiwania
Wyszukiwano:
w słowach kluczowych:  constraint satisfaction problem (CSP)
help Sortuj według:

help Ogranicz wyniki do:
first rewind previous Strona / 1 next fast forward last
EN
Backtrack-style exhaustive search algorithms for NP-hard problems tend to have large variance in their runtime. This is because “fortunate” branching decisions can lead to finding a solution quickly, whereas “unfortunate” decisions in another run can lead the algorithm to a region of the search space with no solutions. In the literature, frequent restarting has been suggested as a means to overcome this problem. In this paper, we propose a more sophisticated approach: a best-first-search heuristic to quickly move between parts of the search space, always concentrating on the most promising region. We describe how this idea can be efficiently incorporated into a backtrack search algorithm, without sacrificing optimality. Moreover, we demonstrate empirically that, for hard solvable problem instances, the new approach provides significantly higher speed-up than frequent restarting.
EN
The article presents the details of the implementation of the concept of a decision support model in the supply chain. To implement the model, the CLP (Constraint Logic Programming) framework called Eclipse was used. The novel way of constraints propagation is discussed, which for this class of problems improves significantly the efficiency of a search for a solution. The most important predicates implementing the model are presented and characterized. Several numerical examples are included to illustrate the implementation of the approach.
PL
W artykule przedstawiono szczegóły implementacji koncepcji modelu wspomagania decyzji w łańcuchu dostaw. Do implementacji modelu wykorzystano środowisko CLP (Programowanie w logice z ograniczeniami) o nazwie Eclipse. Omówiono nowatorski sposób propagacji ograniczeń, który dla tej klasy problemów prowadzi do znacznej poprawy wydajności znajdowania rozwiązania. W artykule przedstawiono i scharakteryzowano najważniejsze predykaty, które służą do implementacji modelu. Jako ilustracje przyjętych założeń i rozwiązań zaprezentowano przykłady liczbowe.
EN
Supply Chain Management (SCM) decisions can be considered at different levels of detail. At a strategic level they apply to the architecture in the supply chain, at the tactical level to transport fleet selection, selection of supply sources and distribution, and at the operational level, to the distribution of supplies and route selection. Many models of decision-making SCM have been developed. These are the linear (LP-linear programming) or mixed (MIP/MILP-Mixed Integer/Linear Integer Programming) models. These models are equipped with a smart form. Although they are well known in the OR (Operation Research) environment, they have significant drawbacks. First of all, they must support only linear constraints. For problems of larger dimensions search for solutions is long and inefficient. This paper proposes a CSP-based decision model for SCM and its implementation in the CLP (Constraint Logic Programming). In addition, it presents a novel way of constraints propagation using the structure of the problem.
PL
Decyzje w zarządzaniu łańcuchem dostaw mogą być rozpatrywane na różnych poziomach szczegółowości. Na poziomie strategicznym dotyczą samej struktury i architektury łańcucha, na poziomie taktycznym wyboru floty transportowej, a na poziomie operacyjnym wyboru tras dostaw itd. Opracowano wiele formalnych modeli zarządzania łańcuchem dostaw. Najczęściej były to modele programowania matematycznego liniowego (LP) oraz całkowitoliczbowego (MILP). Chociaż posiadały struktury dobrze rozumiane w środowiskach (OR-Badań Operacyjnych), posiadały istotne wady. Po pierwsze, mogły zawierać jedynie ograniczenia liniowe. Po drugie nie były efektywne przy większych rozmiarach problemów decyzyjnych. W Artykule zaproponowano model decyzyjny dla łańcucha dostaw oparty na problemie spełnienia ograniczeń (CSP-based) oraz jego implementacji w środowisku programowania w logice z ograniczeniami (CLP). Dodatkowo zaprezentowano nowatorski sposób propagacji ograniczeń wykorzystujący strukturę problemu.
4
EN
The paper aims to present an application of constraint programming techniques for project portfolio scheduling taking into account the imprecision in activity duration and cost. Data specification in the form of discrete a-cuts allows combining distinct and imprecise data, and implementing a constraint satisfaction problem with the use of constraint programming. Moreover using a-cuts, optimistic, pessimistic, and several intermediate scenarios concerning the project scheduling and cash flows can be obtained and considered in terms of different risk levels.
EN
The paper presents a concept for the use of fuzzy modelling and reverse inference to analyze the effects of planned implementation of an ERP system in a medium-sized manufacturing company. The proposed approach allows the possibility of achievement of the effects corresponding to the assumed enterprise targets to be evaluated, and is based on the assignation of suffcient conditions providing these effects defined in the form of selected indicators. The paper also includes a presentation of the results of computer experiments in the field of the determination of the influence of selected fuzzy modelling parameters on the obtained solutions. The suggested concept is dedicated to consultants of ERP software to support decision-making processes.
EN
Constraint Satisfaction Problems typically exhibit strong combinatorial explosion. In this paper we present some models and techniques aimed at improving efficiency in Constraint Logic Programming. A hypergraph model of constraints is presented and an outline of strategy planning approach focused on entropy minimization is put forward. An example cryptoaritmetic problem is explored in order to explain the proposed approach.
7
Content available Modelowanie wielowymiarowych wyrażeń w języku MDX
PL
Artykuł opisuje techniki modelowania zapytań dla modeli analitycznych, wykorzystując język wielowymiarowych wyrażeń (MDX). Przedstawiono narzędzie wspierające wszelkiego rodzaju złożone analitycznie konstrukcje odnoszące się do kostek danych. Zawarto także opis funkcji działających na zbiorach danych jak również zaproponowane przez autora rozwiązania. Wymienione aspekty czynią omawiany język bardzo przydatnym narzędziem wsparcia dla analityków tworzących skomplikowane raporty o dużej złożoności konstrukcyjnej.
EN
The article describes the technique of query modeling for analytical models, using the language of multidimensional expressions (MDX). It is the supporting tool to the all manner of complex query in data cubes. Article include also the description of functions which works on member sets and authors examples of each solutions. All talked aspects of language makes it very useful tool for support analysis in creating complicated reports on large constructional complexity.
PL
W artykule przedstawiono przegląd metody programowania z ograniczeniami w środowisku ILOG OPL. Możliwości tej metody programowania oraz mocne wsparcie dla niej ze strony ILOG Solvera zostały tutaj zwięźle opisane. Omawiane metodyki programowania są bardzo przydatne do rozwiązywania problemów kombinatorycznych, decyzyjnych oraz optymalizacyjnych.
EN
The review of constraint programming method in ILOG OPL environment was presented in this article. Possibilities of this method as well as strong support for it from the ILOG Solver, they became concisely here described. The described methodologies of constraint programming are very useful to the solving of hard combinatorical, decision and optimization problems.
EN
The paper focuses on a selected class of decision problems related with the waste distribution network in SMEs. High efficiency in solving large combinatorial problems offered by implementation of the Constraints Satisfaction Problem (CSP) concept constitute an alternative for the currently available systems based on simulation techniques and operation research. Possible ways of the CSP decomposition as well as possibility of different programming languages application lead then to a problem aimed at searching for a distribution strategy allowing one to interact in an on-line model. The results obtained are implemented in a software package supporting waste distribution network in the SMEs. Illustrative example of the Mozart-based software application is provided.
10
Content available Risk assessement in multi-project environment
EN
Decision making supported by task-oriented software tools plays a pivotal role in a modern enterprise. That is because commercially available ERP systems are not able to respond in an interactive on-line/real-time mode. It opens a new generation of DSS that enable a fast prototyping of production flows in multi-project environment as well as an integrated approach to project execution evaluation. In that context our goal is to provide a knowledge base approach allowing one to be independent on a context or representation of particular data as well as allowing designing an interactive and task-oriented decision support system (DSS). The assumed knowledge base mode of specifying a production system leads to solving a logic-algebraic method (LAM) decision problem. The approach proposed complements the decision system with an additional module (evaluation module) and facilitates searching for possible solutions meeting company production programme execution evaluation criteria. The results obtained are implemented in a software package supporting project management in the SMEs. Illustrative example of the ILOG-based software application is provided.
EN
Solving complex, decision-making problems requires application of decision support systems. The best solving strategy can be defined for each particular problem. The strategy refers to a sequence solving elementary subproblems. The aim of the paper is to present the evaluating criterion to estimate the efficiency of searching strategies. The criterion proposed offers a possibility to estimate the time needed to solve a constraint satisfaction problem. An illustrative example is provided.
first rewind previous Strona / 1 next fast forward last
JavaScript jest wyłączony w Twojej przeglądarce internetowej. Włącz go, a następnie odśwież stronę, aby móc w pełni z niej korzystać.