The service composition problem is a major obstacle that cloud manufacturing platforms must overcome in order to match service providers to consumers such that the requirements of the consumer are met by the service provider. Previous research focuses on supplier selection, quality of service, and semantic web service approaches as a solution to the service composition problem, but rely on subjective aggregations of high-level concepts that provide a distorted perception of the service providers capabilities. Our matching algorithm methodology uses detailed information regarding operation, part, and tooling specific data to create a granular and realistic representation of the service provider. The matching algorithm is validated using a real-world part drawn from the literature and manufacturers in central Pennsylvania. Our findings indicate that the matching algorithm scales linearly as the number of inputs for a parameter increases. The research contribution
of our matching algorithm is significant as its use of detailed machine specific information is novel. Additionally, its inclusion in a cloud manufacturing platform will decrease machine idleness, while providing the customer with an on-demand service that will aid downstream production and design.