Método para análise dos riscos operacionais associados a falhas epidêmicas de novos produtos eletrônicos: uma proposta utilizando redes bayesianas

Abstract

The competition between firms and supply chains, along with the increasing product complexity and existing legal regulations, have been resulting in increased operational risks due to failures of new products. In the electronics industry, especially for the consumer goods segment, the resulting economic risks of such failures may be significant to the profit of firms that design products. This is intensified when the failure rate is high, the so-called 'epidemic failures'. Nevertheless, the assessment of operational risks during the project of new electronic products seems to lack methods to address the uncertainties in a whole integrated approach, taking into consideration the technical and managerial risks, as well as the subjective knowledge of the experts. The present work aims to contribute to the topic, presenting the proposal of a new method for assessing the operational risks associated with epidemic failure of new electronic products. This proposed method was driven by the Design Research method, which enabled the development of a set of artifacts linked through five steps. The main artifact was constructed under the Bayesian Networks approach and it is comprised of a model developed through bibliographic research and interviews with six experts of the electronics industry. Based on the research delimitations, the developed model is composed by 21 constructs, which are interrelated and consider technical and managerial risks associated with Supply Chain, with Product Design, with Verification and Validation tests and with restrictions during the project. The performance evaluation of the method was carried out by a tentative application in a project being implemented at a multinational company established in Brazil. Three sets of potential changes to the project have been identified, for which it was estimated the reduction of operational risks compared to previously established thresholds, as well as evaluated the potential financial results of such changes throughout the product lifecycle. A conclusion is that the method may enhance the firm?s risk assessment process, especially due to the fact that it allows to feedback the probabilistic risk calculations by the record of project evidences. This research, besides contributing with a method proposal to support the new product risk management, indicated potential enhancements to the decision making and knowledge management in project environments.