The era of 5G broadband wireless networks is inextricably connected with the provision of high data rates to mobile users, as well as bandwidth demanding and low latency applications. While large scale deployments of 5G Public Networks are ongoing, enterprises are interested to deploy their own 5G Non Public Networks (NPNs), customized to better serve their specific use cases. To this end, the goal of this paper is to present an architectural approach for cost-efficient 5G Standalone NPN deployment, leveraging cell densification, disaggregated RAN with open interfaces, edge computing and AI/ML-based network optimization. For this purpose, open solutions, such as O-RAN and MANO frameworks for cloud native micro-service deployments are adopted. Furthermore, research, development and deployment challenges are also discussed.