André Machado Caldeira

Ranking of Suppliers for a Steel Industry: A Comparison of the Original TODIM and the Choquet-extended TODIM Methods☆

Abstract This paper presents an application of both the originalTODIM and the Choquet-extended TODIM methods. Both approaches are used here in order to rank suppliers of specified materials and goods quite relevant for a steel industry located in the Brazilian state of Rio de Janeiro. One conceptual advantage from using the Choquet-extended TODIM method is that it can deal with interactions between evaluation criteria. Results from the application of these two methods practically lead to the same ranks. The non-parametric statistic sign test allows concluding that that both methods are indeed equivalent in terms of the final results for this particular application. As a conclusion it can be saidthat the importance of a formal procedure for ranking suppliers basically consists of contributing to greater clarity and a higher learning throughout the ranking process.

Using a Bipolar Choquet Neural Network to Locate a Retail Store

Abstract In this paper, we propose and apply a novel analytical method for solving a class of location problems. The method is based on the use of the bipolar Choquet integral in a neural network for classifying alternative locations. The method is applied to the case study that consists of finding a location for a new store of a major Brazilian retail company. A comparison between the use of the proposed method and the TODIM-FSE method is presented in the paper. The method can also be applied to a situation when one has millions of patterns and new patterns have to be recognized. The paper closes with identifying the proposed method as being efficient and ready to be used with structured and non-structured data systems in order to discover relevant and still unknown information.

Auditing Vehicles Claims Using Neural Networks

Nowadays, fraud is a major enemy of insurance companies. For the total of R

A Fuzzy Inference System to Support Medical Diagnosis in Real Time

28 billion in claims, an estimated R

Sustainability of Brazilian Companies: A Financial Analysis.

7 billions must be fraud. Since the claims represent 59.9% of the premiums paid by the companies, frauds represent 15.0% of them. Therefore, great caution is to be taken in order to detect the frauds and not to pay for these claims. One of the most important detection tools is the audit. However, because it is an expensive service, it is not possible to audit all claims. Based on this caution, the goal of this work is to test some strategies of how to select claims to be audited. The strategies used become more complex, from the first to the fifth, starting with simple thoughts for the first three strategies, and the utilization of logistic models and neural network to estimate the probability of a fraud detection on the fourth and fifth strategies, respectively.

Investments in a New Technological Infrastructure: Decision Making Using the ELECTRE-TRI Methodology

Abstract This paper presents a Fuzzy Inference System created to support medical diagnoses in real time. It minimizes health costs and maximizes health resources based on real data selected by medical groups in the market. This renders support in posting qualified interview decisions through new Customer Risk Mapping. The main objective is to create a foundation of health resources for hospitals administrators based on medical decision procedures thus increasing the capacity for hospitals to absorb new processing techniques and consequently, ensuring higher quality services for their patients. The aim of this system is to delineate patient-risk-factor during a qualifying interview. The automation of this process becomes a benefit by establishing a real-time responses which needs not be applied by a doctor (can be executed by nurses), and, in return, characterizes an opportunity cost for the hospital.

Time Series Forecasting with Neural Networks and Choquet Integral

Abstract This paper compares the Corporate Sustainability Index (CSI index) and the BOVESPA index in the period of 2005 to 2015 to verify the best investment among these two. The data were obtained from BOVESPA and the Brazilian Institute of Geography and Statistics (IBGE) sites. The study has shown that CSI index had an accumulate gain of 108.58% and IBOVESPA‘s accumulate gain was 13.96%. It is also presented a study of the sustainable development index (SDI index) in four fundamental dimensions: environmental, social, institutional and economic.

IDENTIFICAÇÃO AUTOMÁTICA DAS ORDENS DOS MODELOS GARCH UTILIZANDO REDES NEURAIS

Abstract The ELECTRE-TRI multiple criteria decision making method is applied to assist the distribution centres of a Brazilian retail enterprise in order to decide on the investment to be carried out in a new technological structure to be implemented in the central data processing system of the company. This paper illustrates the application of this methodology in a real case study in a major retail chain of stores in Brazil. The practical application of this type of methodology has demonstrated several positive points and huge opportunities for improvement to make it more and more innovative, a very useful tool for decision making in Lojas Americanas.