An Exposed Closed-Loop Model for Customer-Driven Service Assurance Automation

Artificial Intelligence (AI) is widely applied in telecommunications to enable zero-touch automation in network operation and service management. Due to the high complexity, deploying advanced AI mechanisms is not always feasible inside the operator's network domains. Instead, via service exposures, it becomes possible for vertical customers to integrate their external AI solutions with the network and service management system to form a closed loop (CL) and contribute to the automation process. In this paper, we propose an exposed CL model based on service exposure and apply it to automate service assurance tasks like auto-scaling in a network function virtualization (NFV) system orchestrated by ETSI Open Source MANO (OSM). A testbed is built to validate the model. It collects monitoring data from the OSM monitoring module and external monitoring tools. Vertical customers drive and customize their AI solutions to aggregate these data sets and run analytics to detect and predict anomalies prepared for scaling. Preliminary analysis demonstrates the added values of customer-driven monitoring and analysis via the exposed CL.