Linear programming is widely used to solve various complex problems with many variables, subject to multiple constraints. Expert systems are created to provide expertise on complex problems through the application of inference procedures and advanced expert knowledge on facts relevant to the problem. The diet problem is well-known for its contribution to the development of linear programming. Over the years many variations and facets of the diet problem have been solved by means of linear programming techniques and expert systems respectively. In this study the feasibility of combining expert system technology and linear programming techniques to solve a diet problem topical to South Africa, is examined. A computer application is created that incorporates goal programming- and multi-objective linear programming models as the inference engine of an expert system. The program is successfully applied to test cases obtained through knowledge acquisition. The system delivers an eating-plan for an individual that conforms to the nutritional requirements of a healthy diet, includes the personal food preferences of that individual, and includes the food items that result in the lowest total cost. It further allows prioritization of the food preference and least cost factors through the use of weights. Based on the results, recommendations and contributions to the linear programming and expert system fields are presented. / MSc (Computer Science), North-West University, Potchefstroom Campus, 2014

Linear programming is widely used to solve various complex problems with many variables, subject to multiple constraints. Expert systems are created to provide expertise on complex problems through the application of inference procedures and advanced expert knowledge on facts relevant to the problem. The diet problem is well-known for its contribution to the development of linear programming. Over the years many variations and facets of the diet problem have been solved by means of linear programming techniques and expert systems respectively. In this study the feasibility of combining expert system technology and linear programming techniques to solve a diet problem topical to South Africa, is examined. A computer application is created that incorporates goal programming- and multi-objective linear programming models as the inference engine of an expert system. The program is successfully applied to test cases obtained through knowledge acquisition. The system delivers an eating-plan for an individual that conforms to the nutritional requirements of a healthy diet, includes the personal food preferences of that individual, and includes the food items that result in the lowest total cost. It further allows prioritization of the food preference and least cost factors through the use of weights. Based on the results, recommendations and contributions to the linear programming and expert system fields are presented. / MSc (Computer Science), North-West University, Potchefstroom Campus, 2014

Blasting is a critical part of most mining operations. The primary function of blasting is to fragment and move rock. For decades, attempts have been made at increasing the efficiency of blasting to reduce costs and increase production. Most of these attempts involve trial and error techniques that focus on changing a single output. These techniques are costly and time consuming and it has been shown that as one output is optimized other outputs move away from their optimum level. To truly optimize a blasting program, the transfer of explosive energy into individual components must be quantified. Explosive energy is broken down into five primary components: rock fragmentation, heave, ground vibration, air blast, and heat. Fragmentation and heave are considered beneficial components while the remaining are considered waste. Past energy partitioning research has been able to account for less than 30% of a blast’s total explosive energy. The purpose of this dissertation was to account for a greater percentage of the explosive energy available during a blast. These values were determined using measurement techniques not previously applied to energy partitioning research. Four small-scale test series were completed, each designed to isolate individual energy components. Specific energy components measured include borehole chambering, elastic deformation (ground vibration), translational and rotational kinetic energy (heave), and air overpressure (air blast). This research was able to account for 73% of the total explosive energy. Borehole chambering (13%), rotational kinetic energy (25%), translational kinetic energy (5%), and air overpressure (28%) were determined to be the largest components. Prior research efforts have largely ignored rotational kinetic energy and have only been able to offer predictions for the values of borehole chambering and air overpressure energies. This dissertation accounted for a significantly higher percentage of total available explosive energy than previous research efforts using novel measurement techniques. It was shown that borehole chambering, heave, and air blast are the largest energy components in a blast. In addition to quantifying specific energy partitions, a basic goal programming objective function was proposed, incorporating explosive energy partitioning and blasting parameters into a framework that can be used for future energy optimization.

De nos jours, l'entreprise est devenue un monde complexe où se côtoient une multitude de processus plus ou moins formalisés et un ensemble de règles de fonctionnement tacites ou écrites. Selon cette nouvelle organisation, la panne d'un équipement, l'indisponibilité d'une source d'énergie, l'arrêt d'un système automatique, les accidents sont de moins en moins tolérables. Pour ce faire la Fiabilité est maintenant une science qu'aucun concepteur de produit ou d'installation ne peut ignorer. Cependant, l'adoption de la fiabilité limitée sur l'analyse systémique pour la prévision des risques souffre du manque de prise en compte de la variabilité des situations de travail. Cette variabilité se manifeste par la présence de l'opérateur humain comme le postulat de base de l'apparition des événements et des grandes catastrophes des défaillances. En effet, la fiabilité humaine mérite une attention particulière, pourvu que l'opérateur est considéré au coeur du système de travail, il est doublement concerné, donc il est primordial d'appliquer des méthodes pour quantifier son comportement. Dans de telles situations nous proposons d'appliquer un modèle d'optimisation de la fiabilité humaine sous l'initiative de minimisation des risques professionnels par application du modèle programmation mathématique multicritère : Goal Programming avec les fonctions de satisfactions. L'analyse sur le terrain est probante, puisqu'elle permet d'expérimenter le modèle proposé. C'est pourquoi l'étude expérimentale du modèle d'optimisation de la fiabilité humaine est réalisée dans une industrie de traitement de Gaz.

Paskolų teikimas yra viena pagrindinių kredito unijų veiklos krypčių, o pagrindinė ir svarbiausia rizika su kuria jos susiduria yra būtent kredito rizika, todėl būtina ją detaliai analizuoti ir turėti patikimą šių aktyvų valdymo mechanizmą. Taip pat valdant, šią riziką, būtina atsižvelgti į socialinį kredito unijų aspektą. Darbo objektas – kredito rizikos valdymas. Darbo tikslas - išnagrinėti kredito unijų, kaip specifinių finansinių institucijų, veiklos ir kredito rizikos valdymo ypatumus. Pirmoje darbo dalyje nagrinėjami kredito unijų kredito rizikos valdymo ypatumai: analizuojama kooperatyvinių finansinių institucijų kreditavimo specifika bei kredito rizikos ypatybės būdingos kredito unijoms. Taip pat apžvelgiami kredito rizikos vertinimo principai ir pateikiami kredito rizikos valdymo metodai bei jų veikimo ypatybės. Antroje darbo dalyje analizuojama ir apibendrinama kredito rizikos valdymo kredito unijose praktika ir empiriniai tyrimai, apibūdinamos Lietuvos kredito unijų sektoriaus veiklos sąlygos ir formuojama kredito rizikos valdymo Lietuvos kredito unijose tyrimo metodika. Trečioje dalyje tiriami ir vertinami dviejų kredito unijų paskolų portfelio kokybės rodikliai, analizuojama paskolų portfelio koncentracija, remiantis atlikta analize formuojami efektyvūs Lietuvos kredito unijų investiciniai portfeliai, kaip kredito rizikos valdymo instrumentai. Pateikiami siūlymai geresniam kredito rizikos valdymo vystymui Lietuvos kredito unijose. Darbe prieita prie tokių... [toliau žr. visą tekstą] / Lending is one of the main credit union activities, and the most important risks which they face to is credit risk, it is necessary to analyze it in detail and have a sound mechanism of these asset management. As well as the management, this risk is necessary to consider the social aspect of the credit union. The work item - credit risk management. The aim - to examine the operational and credit risk management features of credit union as a specific financial institutions. The first part of the study looked at the credit union theoretical credit risk management features: characterized by cooperative financial institutions lending and credit risk perception of the concept of inherent credit unions. It also reviews the credit risk assessment principles and provides credit risk management techniques and performance characteristics. Secound segment of the paper characterizes situation in Lithuanian credit unions sector and describes researches based on credit risk management in credit unions practice. There also is given research method of credit risk management in Lithuanian credit unions. The work led to the following main conclusions. Credit union is a lending institution which distinguished higher degree of confidence and a wider access to financial services. Profitability is not as much emphasis to be exposed to liberal lending policies, fulfilling the mission and goals. Credit risk assessment before issuing a loan is the most important step in credit risk management process... [to full text]

No description available.

本論文提出數個線性規劃模型建立成長基金的投資組合。目標函數皆以目標規劃方式呈現。第一個模型採用追蹤與成長差距最小的原則。第二個模型改採用大中取小原則。第三個模型則考慮時間因素對於投資組合的影響，修正第一個模型加入時間參數。最後以台灣上市股票市場作為實證分析對象，探討三組模型之表現。 / This thesis presents three linear programming models for selection of the growth portfolio based on historical data. The objective functions of these models are described by goal programming. The first model employs the principle of minimizing the deviation of the value-increasing index. The second model employs the mini-max principle. The third model is derived from the first model and includes the timing effect of historical data during construction of portfolio. The computational results and performance are illustrated by modeling with realistic data from the Taiwan stock market.

Στην παρούσα εργασία παρουσιάζονται εφαρμογές των στοχαστικών διαδικασιών στα λεγόμενα κοινωνικά συστήματα από τη σκοπιά του προγραμματισμού ανθρώπινου δυναμικού (manpower planning). O προγραμματισμός του ανθρώπινου δυναμικού έχει να κάνει με την κατάλληλη τοποθέτηση των μελών του συστήματος στις σωστές θέσεις, σε αριθμούς οι οποίοι εγγυώνται την ομαλή λειτουργία. Αρχικά αναπτύσσουμε το μη ομογενές Μαρκοβιανό σύστημα (ΜΟΜΣ), το οποίο έχει ως βάση του τις Μαρκοβιανές αλυσίδες: η συμπεριφορά του καθορίζεται από την οριακή ή σε πεπερασμένο χρόνο συμπεριφορά μιας μη ομογενούς Μαρκοβιανής αλυσίδας. Το ΜΟΜΣ, είναι ένα μαθηματικό μοντέλο, το οποίο αποτέλεσε μια θεωρία ενοποίησης μέσα σε ένα κοινό πλαίσιο, πολλών γνωστών στοχαστικών μοντέλων προγραμματισμού ανθρώπινου δυναμικού. Στη συνέχεια επικεντρωνόμαστε στον έλεγχο της συμπεριφοράς του μοντέλου. Για το σκοπό αυτό, ορίζουμε μια σχέση για το αναμενόμενο κόστος λειτουργίας και αμοιβών του ΜΟΜΣ. Στη σχέση αυτή δίνουμε τη γενικότερη δυνατή μορφή, έτσι ώστε να περιλαμβάνει ως ειδικές περιπτώσεις αρκετές από τις παραλλαγές των συναρτήσεων κόστους-αμοιβών που υπάρχουν. Τέλος, με τη βοήθεια του προγραμματισμού στόχων (Goal Programming), ελέγχουμε τις ροές του ΜΟΜΣ προκειμένου να επιτευχθεί μια ικανοποιητική συμπεριφορά σύμφωνα με κάποιους στόχους, καθώς το σύστημα περνάει τις τρεις φάσεις λειτουργίας του, δηλ. την παροδική, την ημι-παροδική και τη φάση στατιστικής ισορροπίας. / Manpower planning deals with aspects of human resources management and has been given considerable attention in the last decades. In the attempt to simulate the evolution of a manpower system and predict its future properties, mathematical models were proved to be extremely helpful both for descriptive and optimization purposes. Manpower systems have been modeled in several ways, deterministic or stochastic. The attempt of determine and regulate future structures in a manpower planning system is based mainly on the selection of appropriate recruitment distribution vectors. This effort gives rise to the control problem in mathematical manpower planning. The control of manpower systems has been of considerable concern in recent times. In a series of articles and books beginning back in the early 70's, the problem of finding appropriate recruitment policies was considered, and various mathematical models were developed according to several criteria and practical considerations. In the present work aspiration levels and priorities using goal programming are employed in a NHMS which evolves in three phases, the transient, the semi-transient and the equilibrium phase. The general goal programming framework is used in several variations which depend on the phase, in order to detect appropriate input policies that can achieve a satisfactory trade off between operational cost and target attainability.

No description available.

Submitted by DOUGLAS MIRANDA RODRIGUES null (douglasnet@gmail.com) on 2016-10-20T13:35:28Z No. of bitstreams: 1 Dissertação - Douglas Miranda Rodrigues.pdf: 8525562 bytes, checksum: de3f28242f3d17929bb0eb8ddc912f55 (MD5) / Approved for entry into archive by Juliano Benedito Ferreira (julianoferreira@reitoria.unesp.br) on 2016-10-26T18:00:28Z (GMT) No. of bitstreams: 1 rodrigues_dm_me_guara.pdf: 8525562 bytes, checksum: de3f28242f3d17929bb0eb8ddc912f55 (MD5) / Made available in DSpace on 2016-10-26T18:00:28Z (GMT). No. of bitstreams: 1 rodrigues_dm_me_guara.pdf: 8525562 bytes, checksum: de3f28242f3d17929bb0eb8ddc912f55 (MD5) Previous issue date: 2016-10-10 / Em diversas áreas de trabalho, da Engenharia à Economia, os problemas se apresentam como sendo multiobjetivos, característica que torna complexa a tomada de decisão. Geralmente, estes objetivos são conflitantes e faz-se necessário o uso de técnicas de otimização para a obtenção de melhores resultados. Na presente dissertação serão estudados alguns métodos para a resolução destes problemas, com o objetivo de aplicar métodos de aglutinação em problemas de projetos de experimentos com múltiplas respostas. Deste modo, inicialmente foi realizada uma análise bibliométrica sobre os diferentes métodos utilizados para a resolução destes problemas. A partir disto, foi desenvolvida uma nova abordagem, utilizando a Programação por Compromisso (Compromise Programming – CP) e a Programação por Metas (Goal Programming – GP), bem como diferentes algoritmos (Gradiente Reduzido Generalizado – GRG e a metaheurística do software Optquest) que são usualmente adotados, com comparação de resultados e análise. De modo geral, esta nova proposta apresentou resultados melhores em relação à abordagem tradicional (desirability), qualificando este procedimento como uma alternativa na otimização de múltiplas respostas. / In many areas, from Engineering to Economics, problems present themselves as multiobjective, which makes a decision-making process complex. Generally, these are conflicting objectives, and optimization techniques are necessary to achieve better results. This paper applies agglutination methods in classical problems of design of experiments with multiple responses. A bibliometric analysis was made, and a new approach was developed, using Compromise Programming – CP and Goal Programming – GP, as well as two different algorithms (Generalized Reduced Gradient – GRG and Optquest’s software metaheuristics) with results comparison and analysis. The new proposal presented better results when compared to the traditional approach (desirability), qualifying this procedure as an alternative in multiple responses optimization.