In Cyber Physical Systems humans are often kept in the loop as operators and/or service users. Yet in many cases, humans and machines collaborate and provide services to each other. Research on service models and service composition for CPS exist; however, humans as service providers have not been adequately considered as part of the CPS service composition model. We provide a classification of human-as-a-service in CPS, and we propose a Service Oriented Architecture (SOA) ontology model for the CPS environment as part of the Everything-as-a-Service paradigm. The model considers human characteristics and their dynamics, as a service provider or collaborator with the machine. As the ontology model is an enabler for engineering a self-adaptive CPS with human-machine collaboration as service providers, we describe how a commonly used self-adaptive reference model can be refined to benefit from the vision. We evaluate the ontological contribution against criteria that relates to accuracy, completeness, adaptability, clarity, and consistency. We demonstrate the feasibility of our conceptual reference model using a use case from the medical domain and we show how human-machine service provision is possible.