Abstract: A data store includes item configuration data for a plurality of configurations of an item and inventory mix rate data. A computing device is configured to generate a matrix that combines the item configuration data with the inventory mix rate data such that cells in the matrix include an indication of an inventory mix rate related to an item feature; use the matrix to develop a plurality of neural network inventory turn rate models; and use the turn rate models to associate a value with each configuration in the plurality of configurations.

Abstract: An engine diagnostic system includes a bit-serial based recurrent neuroprocessor for processing data from an internal combustion engine in order to diagnose misfires in real-time and reduces the number of neurons required to perform the task by time multiplexing groups of neurons from a candidate pool of neurons to achieve the successive hidden layers of the recurrent network topology.

Abstract: A method of time-aligning sequential gas concentrations, generated by a non-plug flow real time emissions analyzer, with corresponding plug flow inputs from an emissions source includes selecting a sequence of known plug flow test gas pulses representative of the emissions source, feeding the known plug flow test gas pulses into the analyzer, saving a sequence of blurred test pulses generated by the analyzer, determining a mathematical relationship between the known plug flow test gas pulses and the blurred test pulses, configuring a filter such that the filter corresponds to the mathematical relationship, inputting a subset of the blurred test pulses into the filter, recording a sequence of deconvoluted test pulses generated by the filter, comparing the deconvoluted test pulses with a subset of the known plug flow test gas pulses to generate difference errors, adjusting the filter to minimize the plurality of difference errors, repeating the first inputting step, the first recording step, the comparing step,

Abstract: A process and apparatus for monitoring catalyst conversion activity includes a predictor of feedgas emissions and a predictor of tailpipe emissions, each predictor providing an output for generating a ratio of conversion activity. Each predictor comprises a trained neural network receiving at least one of, and preferably a plurality of, the engine operating condition signals available from an electronic engine control. Preferably, each neural network is trained by inputting accumulated data acquired from performance evaluation of a plurality of vehicles having consistent powertrains but with different degrees of deterioration.

Abstract: A crankshaft based misfire detector for an internal combustion engine detects and identifies misfire over a broad range of engine operating conditions by establishing correction factors to compensate for crankshaft torsional oscillations and static wheel profile errors which corrupt the acceleration signals used to detect misfire. The correction factors are adaptively determined during in-use engine operation and are updated in one manner for most firings but in another manner for any cylinder firing which had the relatively lowest corrected acceleration within an engine cycle.