A semantically constrained Bayesian network for manufacturing diagnosis

The diagnostic problem is posed as recognizing patterns in rejection data and thesubsequent mapping to causes. A new network architecture has been proposedwhich should overcome many of the disadvantages of the existing diagnostictools. The network is based on the authors’ earlier work (Ransing et al. 1995)on representing the causal relationship in the defect-metacause-rootcause form.Although the algorithm is based on the Bayesian analysis, many of the laws ofprobability have been altered to suit the complexities involved. For example, thenotion of conditional probability has been generalized to enable the belief revisioneven in the presence of partial evidence. The inherent presence of the degree ofignorance or uncertainty in the quanti® cation of a relationship has also beenconsidered. Rigorous constraints, again based on the laws of probability, havebeen developed to check the consistency among the network values. The networkis required to be initialized with only a few values or the range for the same andthen a set of globally consistent values is generated automatically and e ciently.Using the most suitable set of consistent values, the diagnosis is performed usingthe generalized Bayesian analysis. The network has been tested for a pressure diecasting process, however, it is generic in nature and can also be applied to othermanufacturing processes