A time-aware Bayesian approach for optimal manufacturing service recommendation in distributed manufacturing environments

With the growing emergence of distributed collaborative manufacturing systems, the utilization of service-oriented manufacturing to offer seamless access to a variety of complex, distributed and dynamic manufacturing resources has become a challenging issue. This paper presents a novel time-aware probabilistic Bayesian approach for recommending a few optimal manufacturing services based on the user preference for an initial manufacturing service. The Bayesian approach operates over a comprehensive, formal representation of manufacturing services, which adds the time-aware probability of satisfied service execution in manufacturing service ontology to take into account the statistical nature of the dynamic manufacturing environment. The Bayesian approach is useful because it infers a few top ranked hypotheses of manufacturing service that have the largest probability as the optimal selection in a specified context of user preference. A prototype system is built and validated with an illustrative example from manufacturing industry to demonstrate the feasibility of the proposed approach for optimal service recommendation.