Detalhes bibliográficos
| Ano de defesa: |
2008 |
| Autor(a) principal: |
Silva, Ribamar Kleber da
 |
| Orientador(a): |
Tanure, José Eduardo Pinheiro Santos
|
| Banca de defesa: |
Figueiredo, Fernando Monteiro de
,
Pessoa, Artur Alves
,
Valente, André Luiz de Carvalho
,
Silva, Kleber Freire da |
| Tipo de documento: |
Dissertação
|
| Tipo de acesso: |
Acesso aberto |
| Idioma: |
por |
| Instituição de defesa: |
Universidade Salvador
|
| Programa de Pós-Graduação: |
Programa de Pós-Graduação em Regulação da Indústria de Energia
|
| Departamento: |
Energia
|
| País: |
BR
|
| Palavras-chave em Português: |
|
| Palavras-chave em Inglês: |
|
| Área do conhecimento CNPq: |
|
| Link de acesso: |
http://teste.tede.unifacs.br:8080/tede/handle/tede/341
|
Resumo: |
The restructuring of the Brazilian electric sector, which began in 1995, opened a new model for regulation of public services of electric energy distribution. Specifically to the distribution s concessionaires, the current regulatory landmark establishes induction mechanisms to the economic efficiency of the monopoly. In this model, the profits of productivity gained by the concessionaire are transferred to the customer in regulatory intervals. In the direction of achieving productivity gains in the management of accounts receivable and in credit policy and collection, the models for assessing the risk of default are based on the classification of the consumer behavior of payment (behavioural scoring), and in forecast of default risk (credit scoring) arise as important tools for decision support. These models are applied in a distribution s concessionaire of electric energy. To this end, data from a sample of consumers of Coelba were used to develop the studies. The statistical techniques used in the construction of the model were: binary logistic regression and cluster analysis. The results of the logistic regression analysis indicate that the model reached a correct classification of default consumers around 84%. The results of the cluster analysis indicate 3 profiles of different clusters of consumers, based on the payment history of the past 6 months. The profile BS1 joined the cases with 91% of the payments made in up to 8 days of bill due date, the best profile of payment. The cases with a situation intermediate were joined in the profile BS2, where 25% of payments are made only on receipt of the invoice, without the use of additional resources for recovery of the default debt. Already the third profile, the BS3, joined the cases with worse profile payment, containing only 6% of payment in up to 8 days of the bill due date and 70% of payment made after the suspension of supply. The studies presented in this paper confirm the contribution of statistical modeling techniques to reduce expenditure with the operations of recovery and the financial losses with bad debtors, through control of the risk of default of the portfolio of consumers |
