Letteratura scientifica selezionata sul tema "Bi-Criteria decision making"

The authors propose a bi-objective two-stage decision-making process to help the marketing team of a company to determine which services make more profit. The decision is based on customer satisfaction measures which are related to the different company services. Thus, they constitute a multi-criteria assessment of the company’s performances. The first stage of the authors’ proposed process is to evaluate the services with respect to certain criteria using a stochastic multi-criteria acceptability analysis. Then, a bi-objective mathematical model is utilized to determine which services are more profitable. An analytical hierarchy process is applied to aggregate the bi-objective model. The applicability and validity of the proposed process is illustrated in a case study.

Z Utility Theory refers to a class of nonlinear utility functions for solving Risk and Multiple Criteria Decision-Making problems. Z utility functions are hybrids of additive and nonadditive (nonlinear) functions. This paper addresses the concepts and assessment methods for the additive part of Z-utility functions for multiple criteria problems that satisfy the efficiency (nondominancy) principle. We provide a decision paradigm and guidelines on how to approach, formulate, and solve decision-making problems. We, also, overview the modeling of decision process based on four types of decision-making styles. For multi-criteria problems, a new definition of convex efficiency is introduced. Also polyhedral efficiency is developed for presenting multi-criteria efficiency (nondominancy) graphically. New double helix quasi-linear value functions for multi-criteria are developed. Two types of double helix value functions for solving bi-criteria (Advantages versus Disadvantages) and also risk problems are introduced: Food–Fun curves for expected values and Fight-Flight curves for expected risk values. Ordinal/Cardinal Approach (OCA) for assessment of additive utility functions is developed. Simple consistency tests to determine whether the assessed utility function satisfies ordinal and/or cardinal properties are provided. We show that OCA can also be used to solve outranking problems. We provide a critique of Analytic Hierarchy Process (AHP) for assessing additive value functions and show that the developed Ordinal/Cardinal Approach overcomes the shortcomings of AHP. We also develop a unified/integrated approach for simultaneous assessment of nonlinear value and additive (multi-criteria) utility functions. These results in an additive utility function that can be concave, convex, or hybrid concave/convex based on the nonlinear value function. Finally, we show an interactive paired comparisons approach for solving nonadditive and nonlinear utility functions for bi-criteria decision-making problems. Several illustrative examples are provided. The paper provides reliable and robust approaches for modeling the utility preferences of heterogeneous economic agents in macro and micro-economics.

TOPSIS (technique for order preference similarity to ideal solution) is considered one of the known classical multiple criteria decision making (MCDM) methods to solve bi-level non-linear fractional multi-objective decision making (BL-NFMODM) problems, and in which the objective function at each level is considered nonlinear and maximization type fractional functions. The proposed approach presents the basic terminology of TOPSIS approach and the construction of membership function for the upper level decision variable vectors, the membership functions of the distance functions from the positive ideal solution (PIS) and of the distance functions from the negative ideal solution (NIS). Thereafter a fuzzy goal programming model is adopted to obtain compromise optimal solution of BL-NFMODM problems. The proposed approach avoids the decision deadlock situations in decision making process and possibility of rejecting the solution again and again by lower level decision makers. The presented TOPSIS technique for BL-NFMODM problems is a new fuzzy extension form of TOPSIS approach suggested by Baky and Abo-Sinna (2013) (Applied Mathematical Modelling, 37, 1004-1015, 2013) which dealt with bi -level multi-objective decision making (BL-MODM) problems. Also, an algorithm is presented of the new fuzzy TOPSIS approach for solving BL-NFMODM problems. Finally, an illustrative numerical example is given to demonstrate the approach.

Abstract Evaluation of the Business Intelligence (BI) competencies of port community systems before they are bought and deployed is a vital importance for establishment of a decision-support environment for managers. This study proposes a new model which provides a simple approach to the assessment of the BI competencies of port community systems in organization. This approach helps decision-makers to select an enterprise system with appropriate intelligence requirements to support the managers’ decision-making tasks. Thirtyfour criteria for BI specifications are determined from a thorough review of the literature. The proposed model uses the fuzzy TOPSIS technique, which employs fuzzy weights of the criteria and fuzzy judgments of port community systems to compute the evaluation scores and rankings. The application of the model is realized in the evaluation, ranking and selecting of the needed port community systems in a port and maritime organization, in order to validate the proposed model with a real application. With utilizing the proposed model organizations can assess, select, and purchase port community systems which will provide a better decision-support environment for their business systems.

Abstract A decision-making process is considered for a firm, in which two coexisting groups of interests pursue different goals. An original model based on a non-neoclassical production function is proposed. The function satisfies the conditions formulated by R. Frisch, which makes it possible to investigate firms operating in the environment far from the perfect competition and pursuing goals other than profit maximization. A two-criteria optimization problem is formulated with the two criteria representing the goals of the groups: maximization of profit and maximization of income generated by the firm with respect to capital and labor. The problem is considered in two variants of the product market, namely the perfect and the imperfect competition. Solutions of the problem are analyzed including the derived Pareto sets. The importance of knowledge about the Pareto set in negotiations between the groups of interests in the firm is illustrated and discussed.

We consider a decision-making problem to evaluate absolute ratings of alternatives from the results of their pairwise comparisons according to two criteria, subject to constraints on the ratings. We formulate the problem as a bi-objective optimization problem of constrained matrix approximation in the Chebyshev sense in logarithmic scale. The problem is to approximate the pairwise comparison matrices for each criterion simultaneously by a common consistent matrix of unit rank, which determines the vector of ratings. We represent and solve the optimization problem in the framework of tropical (idempotent) algebra, which deals with the theory and applications of idempotent semirings and semifields. The solution involves the introduction of two parameters that represent the minimum values of approximation error for each matrix and thereby describe the Pareto frontier for the bi-objective problem. The optimization problem then reduces to a parametrized vector inequality. The necessary and sufficient conditions for solutions of the inequality serve to derive the Pareto frontier for the problem. All solutions of the inequality, which correspond to the Pareto frontier, are taken as a complete Pareto-optimal solution to the problem. We apply these results to the decision problem of interest and present illustrative examples.

L'optimisation combinatoire bi-objectif (BOCO) est un type de prise de décision multicritère avec de nombreuses applications dans différents domaines scientifiques, tels que l'économie, la chaîne d'approvisionnement, les télécommunications, l'informatique et les choix sociaux. Cela implique d'optimiser simultanément deux fonctions objectives (de conflit) dans un ensemble fini et réalisable de vecteurs de décision. Pour BOCO, le concept de solution Pareto-optimale (non dominée) joue un rôle important car il distingue les solutions efficaces et non efficaces. Sur la base de ce concept, les méthodes populaires de résolution de BOCO construisent généralement une représentation de l'ensemble de Pareto qui représente différents compromis entre les objectifs. L'équité proportionnelle est un concept largement étudié dans la littérature qui vise à répartir les services publics pour garantir la justice et l'équité entre les utilisateurs tout en optimisant les performances du système. Concernant les changements proportionnels, il peut être utilisé comme comparaison standard pour deux solutions de BOCO, même lorsque les deux objectifs ont des ordres de grandeur différents. Le but de cette thèse est de proposer un nouveau critère pour sélectionner les solutions Pareto-optimales préférées dans l'ensemble Pareto d'un problème BOCO général utilisant l'équité proportionnelle.À cette fin, nous introduisons le concept de la solution rho-NF (Nash Fairness) pour que BOCO obtienne un certain équilibre de Nash basé sur l'équité proportionnelle. Le facteur rho &gt; 0 est présenté pour refléter l'importance relative entre les deux objectifs. Nous présentons dans un premier temps le concept de solution rho-NF et sa caractérisation. Nous montrons ensuite l'efficacité Pareto des solutions rho-NF. En fait, chaque solution rho-NF est nécessairement une solution Pareto-optimale mais l'inverse n'est pas toujours vrai. Enfin, nous montrons que des solutions rho-NF peuvent être trouvées en optimisant une combinaison linéaire de deux objectifs. Il s'agit d'un aspect crucial dans le développement d'algorithmes efficaces pour identifier les solutions rho-NF.Ensuite, nous présentons les algorithmes exacts pour déterminer les solutions rho-NF pour BOCO. Dans certains cas où l'existence de solution rho-NF n'est pas garantie, nous concevons un algorithme de recherche binaire sur l'espace du rapport entre deux objectifs qui converge selon un nombre logarithmique (de paramètres fixes en fonction des données) d'itérations. Il peut déterminer l'existence de la solution rho-NF et en calculer une dans le cas affirmatif. En revanche, lorsqu'il existe toujours et qu'il peut y avoir de nombreuses solutions rho-NF, nous développons un algorithme récursif de type Newton pour identifier toutes les solutions rho-NF. A chaque appel récursif, on trouve une solution rho-NF en minimisant les combinaisons itératives linéaires de deux objectifs. Notez que le nombre total d'appels récursifs est limité par le nombre de solutions Pareto optimales. De plus, il existe un cas où trouver des solutions rho-NF équivaut à résoudre la programmation fractionnaire classique. Dans ce contexte, nous utilisons la méthode de Newton-Dinkelbach qui est largement appliquée à la programmation fractionnaire. Si deux objectifs sont linéaires, la méthode trouve des solutions rho-NF dans un nombre d'itérations fortement polynomial, quelle que soit la structure de l'ensemble des possibles.Enfin, nous évaluons nos algorithmes à travers diverses instances de BOCO. Les expériences informatiques montrent l'efficacité de nos algorithmes, indiquant leur convergence rapide et leur efficacité dans l'identification de solutions rho-NF. De plus, pour quelques petites instances, nous calculons l'ensemble de Pareto entier pour montrer que l'ensemble de solutions rho-NF est un sous-ensemble strict de l'ensemble de Pareto<br>Bi-Objective Combinatorial Optimization (BOCO) is a type of multiple-criteria decision-making with many applications in different scientific areas, such as economics, supply chain, telecommunications, computer science, and social choice. It involves optimizing simultaneously two (conflict) objective functions in a finite feasible set of decision vectors. For BOCO, the concept of Pareto-optimal (non-dominated) solution plays an important role as it distinguishes between efficient and non-efficient solutions. Based on this concept, popular methods for solving BOCO usually construct a representation of the Pareto set that represents different trade-offs between the objectives. Proportional fairness is a widely studied concept in the literature that aims to distribute utilities to ensure fairness and equity among users while optimizing system performance. Related to the proportional changes, it can be used as a standard comparison for two solutions of BOCO, even when the two objectives have different orders of magnitude. The goal of this thesis is to propose a novel criterion for selecting preferred Pareto-optimal solutions in the Pareto set of a general BOCO problem using proportional fairness.For this purpose, we introduce the concept of rho-NF (Nash Fairness) solution for BOCO achieving some Nash equilibrium based on proportional fairness. The factor rho &gt; 0 is presented to reflect the relative importance between the two objectives. We first present the rho-NF solution concept and its characterization. We then show the Pareto efficiency of rho-NF solutions. In fact, each rho-NF solution is necessarily a Pareto-optimal solution but the inverse is not always true. Finally, we show that rho-NF solutions can be found by optimizing a linear combination of two objectives. This is a crucial aspect in developing efficient algorithms for identifying rho-NF solutions.Subsequently, we present the exact algorithms to determine the rho-NF solutions for BOCO. In some cases where the existence of rho-NF solution is not guaranteed, we design a binary search algorithm on the space of the ratio between two objectives which converges in a logarithmic (of fixed parameters depending on the data) number of iterations. It can determine the existence of the rho-NF solution and compute one in the affirmative case. In contrast, when there always exists and may be numerous rho-NF solutions, we develop a Newton-like recursive algorithm to identify all the rho-NF solutions. At each recursive call, we find a rho-NF solution by minimizing iteratively linear combinations of two objectives. Notice that the total number of recursive calls is bounded by the number of Pareto-optimal solutions. Furthermore, there exists one case where finding rho-NF solutions is equivalent to solving classical fractional programming. In this setting, we use the Newton-Dinkelbach method which is broadly applied to fractional programming. If two objectives are linear, the method finds rho-NF solutions in a strongly polynomial number of iterations, regardless of the structure of the feasible set.Finally, we evaluate our algorithms through various instances of BOCO. Computational experiments show the effectiveness of our algorithms, indicating their rapid convergence and efficiency in identifying rho-NF solutions. Furthermore, for some small instances, we compute the whole Pareto set to show that the rho-NF solution set is a strict subset of the Pareto set

Business Intelligence (BI) applications are consulted by their users on a daily basis. BI information obtained assist users to make business decisions and allow for a deeper understanding of the business and its driving forces. In a mining environment companies need to derive maximum benefit from BI applications, therefore these applications need to be used optimally. Optimal use depends on various factors including the usability of the product. The documented lack of usability evaluation guidelines provides the rationale for this study. The purpose is to investigate the usability evaluation of BI applications in the context of a coal mining organization. The research is guided by the question: What guidelines should be used to evaluate the usability of BI applications. The research design included the identification of BI usability issues based on the observation of BI users at the coal mining organization. The usability criteria extracted from the usability issues were compared and then merged with general usability criteria from literature to form an initial set of BI usability evaluation criteria. These criteria were used as the basis for a heuristic evaluation of the BI application used at the coal mining organization. The same application was also evaluated using the Software Usability Measurement Inventory (SUMI) standardised questionnaire. The results from the two evaluations were triangulated to provide a refined set of criteria. The main contribution of the study is the heuristic evaluation guidelines for BI applications (based on these criteria). These guidelines are grouped in the following functional areas: visibility, flexibility, cognition, application behaviour, error control and help, affect and BI elements.<br>Information Science<br>M.Sc. (Information Systems)

AbstractIn today’s global business context, data has played a critical role in ensuring accurate and appropriate decision making in manufacturing organisations. Despite the huge pool of information (i.e. data) generated by consumers, repair or maintenance shops, manufacturing job shop, scientific society on various products, which could be deployed by manufacturers in eliciting vital information towards achieving sustainable product design and development, only few manufacturers are making use of this data to generate wisdom required for sustainable manufacturing. This act is caused by lack of appropriate systems capable of integrating the available data and make wise inferences that will result in a competitive advantage of a specific organisation over its competitors. In light of this, the aim of this study is to establish a suitable data analytic platform that could be used to sort, classify and integrate data required to generate wisdom vital for sustainable manufacturing. In order to achieve this, Analytical Hierarchy Process (AHP) was deployed to appraise various alternative data analytical platforms such as Python, Apache Spark, Qlik View, Power BI, Tableau, KNIME, Excel, Talend, Rapid Miner and Statistical Analysis System (SAS) using various criteria such as Data Format, Availability, Interface, Programming Intensity, Data Science Knowledge Intensity and Capabilities. The result of this decision analysis and selection exercise, revealed that KNIME data analytic platform, with the most important decision criterion; data science knowledge intensity, and a cumulative assessment score of 80.80 is the appropriate data analytic platform that manufacturers should use to generate a knowledge advisor vital for sustainable manufacturing and product development.

AbstractCurrently, levels of production and disposal for communication devices such as smartphones are continuing to increase. In the life cycle of a smartphone, the majority of greenhouse gas (GHG) emissions are generated in the material production stage. To recover the GHG emissions from end-of-life (EOL) products such as smartphones, manufacturers have to recycle EOL products. However, smartphones on the market undergo little recycling because costs related to recycling, transportation, and facilities are very high. Therefore, the decision maker (DM) has to design a reverse supply chain network for collecting EOL products from users and transporting them to recovery or disposal facilities not only environmentally friendly but also economically feasible. This study applies a bi-objective reverse supply chain network design to material-based GHG volumes and related costs applying a multi-criteria decision-making methods as linear physical programming (LPP) to design a reverse supply chain network in the case of smartphones. First, the reverse supply chain network is modeled for recycling EOL smartphones, and a case study based on literatures and life cycle assessment are prepared. Next, the objective functions are set and formulated to minimize the total volume of material-based GHG volume and the total cost using LPP and integer programming. Finally, numerical experiments on the reverse supply chain are conducted and evaluated.

Business intelligence (BI) is an integrated set of tools used to support the transformation of data into information in order to support decision making. Among different functionalities, reporting plays a significant role that provides information to its readers to make better decisions. BI lacks user-specific reporting to the different levels of users of an organization. Different users require different kinds of reporting with respect to different requirement (criteria) in an organization. A multi-criteria reporting (MCR) finds the suitable information to suitable user based on the multiple conflicting preferences of a user. Technique for order preference by similarity to ideal solution (TOPSIS) is the most popularly applied multi-criteria decision-making (MCDM) technique selected to identify different levels of user preference for MCR. Banking business is considered as a case study to identify user preference for MCR. This research also designs evaluation metrics for TOPSIS.

The optimization of the maintenance decision making can be defined as an attempt to resolve the conflicts of decision situation in such a way that variable under the control of the decision maker take their best possible value. One of the most important controllable parameters is the time interval between maintenance. Most of the researchers have kept the fact that whenever the suitable maintenance interval is reached, the system is replaced with the original one. However the improvement of a system life not only depends on the replacement of deteriorated components, but also on the effectiveness of the maintenance. Taking care about this fact, the effects of maintenance of a multi-component system by combining the three main different PM actions, namely (1a), (1b) and (2p)-maintenance actions. Thus, the main purpose of an effective maintenance program is to present a technique for finding the optimal maintenance interval for the system by considering the multiple goals of the organization viz. maximum availability, minimum maintenance cost.

Surveys are an effective way to gather large amounts of information from many respondents. They are flexible and can include a number of topics and subtopics that are selected to focus on the unique aspects of a particular school or district. This flexibility means that questionnaires that address multiple issues can be combined into one survey. There are many good reasons to compose a survey from existing questionnaires and scales. Most existing instruments were developed by experts and tested and improved over the years. Many of them are free, open for anyone to use, and readily available to schools interested in including them in their monitoring systems. Many questionnaires can be found in special compendiums that store large numbers of instruments, often relevant to a particular area (see Box 8.1). Schools choose from available questionnaires and scales based on multiple criteria, the most important being their relevance to school needs and what school leaders want to know. These decisions pertain to the content of the instrument but also to the language used. Certain wording, for example, is not useful with younger students or might not communicate well with non– English- speaking parents. Some instruments were designed to cover a certain topic in depth, making it relatively long. Schools interested in mixing several instruments to cover many topics may prefer a shorter questionnaire that fits with others in the same survey. It’s also important to give surveys on a regular, consistent basis. Giving a survey only once won’t provide school leaders with the information they need to improve school climate over time. A true monitoring system will include administering the sur­vey on an annual or bi- annual basis so that educators can tell whether issues are getting better or worse, which groups of students are affected more than others, and how their schools compare to others in the district or across the state. At times, districts and schools may need and want to create their own questionnaires or at least add modules to existing ones. This may happen, for instance, when the school is interested in monitoring the effects of certain practices implemented by the school.