Multiple Criteria Model Research Articles

Anaemia, iron deficiency and vitamin A status among school-aged children in rural Kazakhstan

To investigate the prevalence of anaemia and iron deficiency and vitamin A status among school-aged children in rural Kazakhstan and identify factors associated with anaemia in this population. A cross-sectional design. School-aged children in rural Kazakhstan. Socio-economic and anthropometric information was collected from 159 school-aged children living in the Kzyl-Orda region of Kazakhstan. Blood samples were collected and the concentrations of haemoglobin (Hb), serum iron, serum ferritin (SF), erythrocyte protoporphyrin (EP), serum retinol and beta-carotene, total iron binding capacity (TIBC), transferrin saturation (TS) and other haematological indices were measured. Among the 159 children, the prevalence of anaemia and iron deficiency defined by the multiple criteria model (SF, TS and EP) was 27% and 13%, respectively. Nine per cent had iron-deficiency anaemia and 21% had serum retinol value < 1.05 micromol l(-1). Mean SF and serum iron concentrations and TS were significantly lower in anaemic children than in their non-anaemic peers, while TIBC and EP were significantly higher in children with anaemia. Hb was significantly correlated with serum iron and retinol concentrations. Serum retinol and SF concentrations and mean corpuscular volume were significantly correlated with Hb by multiple regression analysis. Anaemia among school-aged children in rural Kazakhstan appears to be related to iron indices and vitamin A status.

A formulation of portfolio selection problem with multiple criteria

AbstractPortfolio‐type problems arise in many decision‐making situations. In this paper, we consider these problems in the context of multiple criteria model that takes into account two types of criteria, which are used in decision maker's preferences: criteria of the first type are used to characterize both the entire portfolio and its individual elements, whereas criteria of the second type are solely used to evaluate the portfolio as a whole but not its elements. When formulating our multiple criteria model, we assume that criteria of the first type are quantitative and that they represent a higher priority for the decision maker compared to criteria of the second type. While we acknowledge the existence of other approaches, in this paper we focus on those portfolio selection problems that match the above structure of the criteria set. After formulating the portfolio selection problem, we offer a two‐phase portfolio selection procedure that allows to choose an optimal portfolio by operating in two separate criteria spaces. Copyright © 2006 John Wiley &amp; Sons, Ltd.

Multiple Criteria Approach to Solving Problems of Metal Building-Up by Welding

The problem of process optimizing of metal building-up by welding is investigated. A multiple criteria model with four objectives is suggested. To model the process parameters, a planned experiment has been carried out. The interactive satisfying trade-off method (STOM) of Nakayama is used to solve the model.

Use of Combined Measures from Capillary Blood to Assess Iron Deficiency in Rural Kenyan Children

Community-based surveys of iron deficiency (ID) require simple, accurate methods that can be used in remote areas. The objective of this study was to assess iron status in rural Kenya using "field-friendly" methods for capillary blood, including an improved dried blood spot assay for transferrin receptor (TfR). A single finger stick was used to obtain capillary blood from 275 school-age children. Whole blood was applied directly to filter paper, dried, and later analyzed for TfR, as well as C-reactive protein (CRP), an acute-phase protein that serves as a general marker of inflammation. Capillary blood was also used to measure hemoglobin (Hb) concentration and the ratio of zinc protoporphyrin to heme (ZPP:H). The Hb concentration alone provides the lowest estimate of the prevalence of ID (8.0%). Because ZPP:H is reported to be elevated in the presence of inflammation, we constructed a preliminary diagnostic model based on elevated ZPP:H and normal CRP level, estimating the prevalence of ID at 25.9%. When TfR is added to a multiple criteria model (elevated ZPP:H in the absence of elevated CRP and/or elevated TfR level) the prevalence of ID is estimated to be 31.2%. This study demonstrates the diagnostic utility of combining TfR with other indexes of iron status, enabling the detection of ID in both the presence and absence of infection. Furthermore, this study is the first field application of TfR blood-spot methods, and it demonstrates their feasibility in remote field settings.

Multiple criteria decision making and risk analysis as risk management tools for power systems planning

Uncertainties in power systems planning are getting more important nowadays due to the liberalization of the electricity industry and the increasing concern for the environmental impact of electricity generation. This paper presents an electricity planning model which deals with uncertainty and its associated risk at two levels: at the first level, by minimizing environmental risk through a multiple-criteria model; and at the second level, by performing a risk analysis consistent with the multiple criteria model used before, and which applies classical decision rules for selecting the best planning strategy under uncertainty. Results show that the incorporation of additional criteria produce much more flexible and efficient strategies, which greatly reduce environmental risk at a little cost increment, while the risk analysis process selects flexible and robust strategies for the scenarios analyzed.

Multiple Criteria Decision-Making and Risk Analysis as Risk Management Tools for Power Systems Planning

Uncertainties in power systems planning are gaining importance due to the liberalization of the electricity industry and the increasing concem for the environmental impact of electricity generation. This paper presents an electricity planning model that deals with uncertainty and its associated risk at two levels: at the first level, by minimizing environmental risk through a multiple-criteria model; at the second level, by perfonning a risk analysis consistent with the multiple criteria model used before and which applies classical decision rules for selecting the best planning strategy under uncertainty. Results show that the incorporation of additional criteria produce much more flexible and efficient strategies, which greatly reduce environmental risk at a little cost increment while the risk analysis process selects flexible and robust strategies for the scenarios analyzed.

A Multi-Attribute Negotiation Support System with Market Signaling for Electronic Markets

Despite the rapid growth of technology and Internet-based markets, many of the current systems limit themselves to price as the single dimension variable and offer, if at all, only minimal negotiation support to the consumer. In the real world, commercial transactions take into account many other parameters both quantitative and qualitative such as product quality, speed, reputation, after sales service, etc. This paper discusses how these multiple attributes can be captured to augment standard negotiation processes in order to support electronic market transactions. Using a combination of utility theory and multicriteria decision-making, we propose heuristic algorithms to discover potential trades. In addition, the approach is included within a larger framework that incorporates market-signaling mechanisms. This not only allows for the systematic evolution of negotiation positions among buyers and sellers but can ultimately lead towards improving both market transparency and efficiency. To illustrate the multiple criteria model coupled with the dynamic market signaling framework, we report in this paper the implementation of a Web-based clearinghouse that serves the real estate market.

Inequality measures and equitable approaches to location problems

Location problems can be considered as multiple criteria models where for each client (spatial unit) there is defined an individual objective function, which measures the effect of a location pattern with respect to the client satisfaction (e.g., it expresses the distance or travel time between the client and the assigned facility). This results in a multiple criteria model taking into account the entire distribution of individual effects (distances). Moreover, the model enables us to introduce the concept of equitable efficiency which links location problems with theories of inequality measurement. In this paper special attention is paid to solution concepts based on the bicriteria optimization of the mean distance and the absolute inequality measures. The restrictions for the trade-offs are identified which guarantee that the bicriteria approaches comply with the concept of equitable efficiency. These results are further generalized to bicriteria approaches not using directly the trade-off technique.

Multiple criteria linear programming model for portfolio selection

The portfolio selection problem is usually considered as a bicriteria optimization problem where a reasonable trade-off between expected rate of return and risk is sought. In the classical Markowitz model the risk is measured with variance, thus generating a quadratic programming model. The Markowitz model is frequently criticized as not consistent with axiomatic models of preferences for choice under risk. Models consistent with the preference axioms are based on the relation of stochastic dominance or on expected utility theory. The former is quite easy to implement for pairwise comparisons of given portfolios whereas it does not offer any computational tool to analyze the portfolio selection problem. The latter, when used for the portfolio selection problem, is restrictive in modeling preferences of investors. In this paper, a multiple criteria linear programming model of the portfolio selection problem is developed. The model is based on the preference axioms for choice under risk. Nevertheless, it allows one to employ the standard multiple criteria procedures to analyze the portfolio selection problem. It is shown that the classical mean-risk approaches resulting in linear programming models correspond to specific solution techniques applied to our multiple criteria model.

On the distribution approach to location problems

While making location decisions, the distribution of travel distances among the service recipients (clients) is an important issue. It is usually tackled with the minimax (center) or the minisum (median) solution concepts. Both concepts minimize only simple scalar characteristics of the distribution: the maximal distance and the average distance, respectively. In this paper, all the distances for the individual clients are considered as a set of multiple uniform criteria to be minimized. This results in a multiple criteria model taking into account the entire distribution of distances. Our analysis of the multiple criteria problem focuses on the symmetrically efficient solutions which comply with minimization of distances as well as with impartial consideration of the clients. Various solution concepts generating symmetrically efficient location patterns are analyzed. Finally, the reference distribution approach is developed as an interactive technique which enables us to identify a satisfactory symmetrically efficient location pattern by evolving a reference (target) distribution of distances.

MULTI-CRITERIA ASSESSMENT OF ECOLOGICAL PROCESS MODELS

The Pareto Optimal Model Assessment Cycle (POMAC), a multiple-criteria model assessment methodology, is described for exploring uncertainty in the relationships between ecological theory, model structure, and assessment data. Model performance is optimized to satisfy, simultaneously, each component of a vector of assessment criteria (model outputs), rather than the usual procedure of optimizing performance with respect to a single criterion. Pareto Optimality is used to define the vector optimization. The Pareto Optimal Set reveals which combinations of assessment criteria the model can satisfy simultaneously. Binary interval error measures, which classify whether a parameterization result is within an acceptable range of values, are defined for each criterion. Their use masks small differences in the performance of different parameterizations, allowing the Pareto Optimal Set to reveal conflicts in ability to achieve simultaneously different collections of criteria. POMAC improves the researcher’s ability to detect deficiencies and locate their sources. It is more stringent and informative than traditional model assessment procedures because it uses multiple criteria without weighting and aggregating them. The Pareto Optimal Set reveals the presence of deficiencies through the model’s inability to satisfy all the criteria simultaneously. POMAC then guides the researcher in locating deficiencies in: inadequate selection of component ecological hypotheses underlying the model, inadequate mathematical representations of these hypotheses, inadequate parameterization, poor selection and formulation of the assessment criteria, or combinations of these. In an example, POMAC is applied to the spatially explicit canopy competition model WHORL using ten assessment criteria. Each criterion was selected to provide information on different aspects of WHORL’s functioning: three stand height distribution criteria, three crown morphology criteria, and four criteria focusing on stand competition’s characteristic differentiation of growth rates. The Pareto Optimal Set was generated using simulated evolution optimization. POMAC revealed deficiencies in both the model structure and its assessment criteria, leading to an improved model and better understanding of its effective domain.

A nonlinear multiple criteria model base for the problem of balancing caseload risk in probation departments

This paper presents a combined statistical model for probationer risk classification and nonlinear multi-criteria programming resource allocation model for workload balancing. A statistical model of individual probationer failure was used to classify individuals as high, medium or low risk for probation purposes. The proportion of probationers in each risk category was used as an input into the integer, nonlinear resource allocation model where the decision variables were the number of visits to probationers by type of visit and risk classification.

Multiple criteria modelling and ordinal data: Evaluation in terms of subsets of criteria

In an earlier article an ordinal multiple criteria model was presented in which each member of a set of alternatives was given an evaluation on each member of a set K of criteria. In this paper we extend this concept to the situation where each alternative i can be assessed in terms of only a subset Ki of K. Various models are presented for dealing with this partial criteria case and the pros and cons of these models are discussed.

Reasoning on infeasibility in distributed collaborative computing environment

Collaborative computing is a new communication technology that makes it possible to extend model formulation, management, and analysis into a geographically distributed group environment. The forms of communication may vary from asynchronous hypermedia teamwork via the Internet to multipoint desktop video conferencing. The latter presents the maximal potential for integrating shared quantitative model analysis with real-time/asynchronous geographically distributed electronic meetings. Infeasibility diagnosis and reasoning on conflict resolution are known to be important parts of an evolving approach to linear programming model analysis. In the new distributed environment, several specific decision support issues related to infeasibility analysis emerge. One is the learning mechanism for capturing group knowledge on infeasibility resolution that is generated during the collaborative sessions of model analysis. The other new decision support component is a coordination protocol that is capable of linking individual activities and software transactions for the support of group reasoning on infeasibility analysis. This paper addresses these issues on the basis of committee models for infeasibility resolution and the neural network approach. Examples of modeling cases are based on experiments using a multiple criteria model for support of resource allocation in a distributed electronic meeting, and the case of model-based diagnosis that the group is involved in.

A multiple-criteria decision model with ranked inputs

A multiple-criteria decision model with ranked inputs

A multiple criteria model for new telecommunication service planning

The strategic planning of telecommunications networks, namely concerning the introduction of new service offerings, is a problem which involves different and conflicting aspects. Interactive multiple criteria methods are well suited to tackle these complex problems because they allow the intervention of the decision maker in the solution search process. This paper aims to present a multiple criteria linear programming model for modernization planning of telecommunications networks. The main features of a decision aid computational tool, the fundamental component of which is a method base which takes advantage of the combination of different types of interactive procedures, will be described. Its potentialities to provide decision support in the evaluation of network modernization strategies will be illustrated through an example.

Project Planning and Prioritization in the Social Services–an OR Contribution

This paper describes a study carried out for the Social Services Department of a UK County Council towards the development of a system for the evaluation and prioritization of project-based work. There are three major components of the study; an activity analysis to investigate how managers in the department spend their time, with a view to establishing a time budget for the project work; a pilot study to introduce a simple system for project planning and workload balancing with a specialist team; and the development of a multiple criteria model for the evaluation and prioritization of projects. We envisaged a Decision Support System which integrated all of these elements; however, this was never developed; instead, the resultant benefits were of a very different nature. We reflect on the reasons for this and the implications for the success of the study.

A problem identification taxonomy for classification and automated formulation of multiple objective linear programming models

This paper presents a knowledge-based Decision Support System (DSS) for classification, formulation and solution of multiple objective linear programming problems. The authors propose a generic taxonomy which is used to classify commonly encountered types of linear programming problems. Classification of the problem, vis-a-vis the proposed taxonomy, is based on the interaction with the user to determine the attributes which best describe the context or setting of the problem. A total of twenty-four problem types are included in the taxonomy. Following classification, a problem type-specific rule base is invoked to assist the user in constraint formulation. A product blending linear programming problem is used to demonstrate this component of the system since these types of problems typically include more varied constraints, including ratio as well as additive types. A second rule base is invoked for formulation of the multiple criteria objective function, model solution and sensitivity analysis. An initial goal prioritization scheme is obtained by use of the Analytical Hierarchy Process [17]; the optimal goal ordering is obtained by interaction with the user in the form of pairwise attribute value tradeoffs. The system developed is intended for use by the OR-naive user who is more familiar with the content of a problem than he/she is with the mathematical tools needed to formulate and solve such models. The system is a model management tool designed to interpret user inputs and translate those inputs into a solvable multiple objective LP. This interface alleviates the technical burdens of content specification and solution. The approach expands previous formulation tools, such as those based on natural language processing [8], to a broader range of problem types in a multiple criteria environment. The system was implemented on a personal computer using VP-Expert, BASIC and LINDO, and is demonstrated on a multiple objective blending problem. The ability of the approach to accurately classify LP problems was tested on thirty-six subjects. Results suggest that correct classification of problems was more likely to occur when the system was used.

A multiple criteria model for the allocation of data files in a distributed information system

This paper addresses a multiple objective optimization model to determine the file allocation and query routing assignment in a distributed information system. The problem is formulated as a zero-one integer nonlinear programming problem with multiple objectives. The optimization problem under consideration is shown to be NP-hard. We adopt an iterative improvement procedure, which gives the Pareto optimal solution. We further illustrate our methodology with a small sample system.

A multiple criteria model for audit planning decisions*

Abstract. The auditor must trade off potential losses from both ineffectiveness (failure to detect client errors) and inefficiency. However, auditing standards do not specify how the auditor is to achieve an appropriate balance between these two conflicting objectives. This paper presents a multiple criteria decision‐making (MCDM) model of an audit planning decision in which an effectiveness vs. efficiency trade‐off is required. The key advantage of the MCDM model developed here is that an optimal solution is obtained without requiring ex ante specification of a loss function by the auditor. Thus, the model addresses a previously recognized problem (e.g., Menzefricke, 1984) that loss functions may vary depending on characteristics of the audit. In this paper, we develop a model of the auditor's sampling plan selection problem incorporating multiple decision criteria, use the model under a variety of simulated audit conditions, and compare our results with previous research.Résumé. Le vérificateur doit parvenir à un compromis entre les pertes pouvant être occasionnées à la fois par l'inefficacité (non‐détection d'erreurs commises par l'entreprise cliente) et l'inefficience. Les normes de vérification ne précisent cependant pas comment le vérificateur doit parvenir à un équilibre approprié entre ces deux objectifs conflictuels. Les auteurs proposent un modèle de dècision à critères multiples qui s'applique à la décision de planification de la vérification exigeant un compromis efficacité‐efficience. Le principal avantage de ce modèle est qu'il permet d'obtenir une solution optimale sans exiger la formulation ex ante par le vérificateur d'une fonction de perte. Le modèle porte done sur une question sur laquelle se sont déjà penchés les chercheurs (Menzefricke, 1984, par exemple): les fonctions de perte peuvent varier selon les caractéristiques de la vérification. Les auteurs élaborent ici un modèle qui s'applique au problème de sélection par le vérificateur d'un plan d'échantillonnage incorporant des critères de décision multiples. Ils utilisent le modèle en simulant diverses situations de vérification et comparent les résultats obtenus avec ceux des travaux précédents.