Abstract: A business process modeling framework is used for data quality analysis. The modeling framework represents the sources of transactions entering the information processing system, the various tasks within the process that manipulate or transform these transactions, and the data repositories in which the transactions are stored or aggregated. A subset of these tasks is associated as the potential error introduction sources, and the rate and magnitude of various error classes at each such task are probabilistically modeled. This model can be used to predict how changes in transactions volumes and business processes impact data quality at the aggregate level in the data repositories. The model can also account for the presence of error correcting controls and assess how the placement and effectiveness of these controls alter the propagation and aggregation of errors.

Abstract: Steel making is a batch production process in which batches of 250 tons, called charges or heats, are grouped together in a single production run, called a cast. Because of fixed costs and downtime associated with cleanup and setup between casts, a cast of less than three charges is likely to be inefficient. As a result, short-run orders (such as, for example, orders of less than 750 tons) may present efficiency problems. The present invention transforms the operation from a make-to-order model to a make-to-stock model, enabling orders to be clustered based on considerations such as grade and geometry to maximize the number of clusters that satisfy the cast-size constraint.

Abstract: A computer implemented process is provided for fast and accurate evaluation of the performance characteristics of the periodic-review (s,S) inventory policy with complete back ordering. This policy has an underlying stochastic process that is a renewal process. The method provides a novel computer implementation of a fast and accurate way to compute approximations of the renewal function. In order to overcome the computational problems in evaluating renewal functions numerically, an approximation scheme has been devised whereby the renewal function of the truncated normal distribution can be characterized by two parameters: (1) its coefficient of variation, and (2) the point at which the function needs to be evaluated. This approximation is derived in two stages. In the first stage, a class of rational polynomial approximations are developed to the renewal function, called Pade approximants.