Self-Optimizing Algorithm for Real-Time Problem Resolution Using Historical Data

摘要

A self-optimizing algorithm for real-time problem resolution using historical data. Upon receiving failure symptom characteristics for a product or process failure, the algorithm queries historical failure data to locate historical failure symptoms and corrective actions matching the failure symptom characteristics. If a total number of the historical corrective actions identified meets a minimum match threshold, the algorithm selectively prunes a failure symptom characteristic having the lowest priority level to form an adjusted search query. The algorithm may repeat the querying, identifying, and determining steps using the adjusted search query until the total number of historical corrective actions identified meets the minimum match threshold. Once the threshold is met, the algorithm sorts the historical corrective actions to form a list of recommended corrective actions for the failure symptom characteristics and provides the list of recommended corrective actions to an end user.