Edge-to-Business Value Chain Delivery via Elastic Telemetry of Cyber-Physical Systems

The Cyber-Physical Systems (CPS) brought great computational advancements to transform the computing paradigm into a living ecosystem from edge infrastructures to business value propositions. Such advancements form a collaboration network among various entities, services and things which we call Edge-to-Business (E2B) value chain. However, along with the rise of E2B evolution, there is a need for blueprints on how to define your business models, design principles on how to build your applications, and a protocol on how to generate revenues from your ecosystem. The three-part work plan of the this thesis is to: (i) discover such blueprints, (ii) provide robust design patterns; and (iii) develop fairly flexible metering mechanisms to achieve economies of scale within this virtual value chain. In part one, we first identify the value creation models in the CPS application domains of cloud manufacturing, open government data and health-care. To be more specific, we incorporate product topology templates as a marketable entity for cloud manufacturing service. Next, we propose a novel abstraction unit called Gov. Data Compute Unit (DCU), so that governments are able to feed developers with formalized, structured and programmable data resource units rather than just data catalogs. For the last domain, we develop a solution for the semantic sensor data retrieval from the edge device layer. This enriches the quality of raw sensed data and makes it more interpretable. The second part of the study deals with developing eight IoT design patterns as computational constructs for engineering edge applications. Such abstract solutions comprise core design principles through CPS life cycle. Our patterns cover applications’ iteration cycles like provisioning, edge deployment pipeline, dynamic configurations, constrained application messaging, in-device data processing and wearable façade creation. To complete the story, a telemetry protocol is developed at part three to capture the generated value of CPS applications. We offer the Diameter of Things (DoT) protocol that implements a real-time metering in the case of prepaid and pay-per-use economic models. The protocol handles time-based and event-based telemetry patterns. Developed approaches and solutions were reasonably validated to prove the validity and utility of our work. The techniques developed in this thesis give a better understanding of the E2B value chain and can be extended to deliver more diverse business models, more edge design patterns as well as more complex metering models.