Uninterrupted and high-quality data communication is one of the most important prerequisites for implementing OT (Operation Technology)-grade private wireless like LTE and 5G NR for industrial automation. Such networks are expected to enable operations 24/7 in the covered area with given network service level agreements (SLA), e.g., bandwidth and latency requirements that are use-case specific. One of the main challenges in designing 5G industrial networks is that the network conditions might change drastically over a short time due to a variety of factors, including moving parts, changes in the factory floor layout, and stored goods temporarily affecting the wireless signal. Motivated by the complexity and variety of current factory use cases, this proposes a new platform enabling continuous testing and SLA management of wireless networks that enables: i) the creation and management of SLA definitions for each connected device through a custom-built SLA compiler, ii) all the automation needed to continuously test the network continuously, according to the defined SLAs, and iii) an adaptation engine that is capable of reallocating software resources on the edge (e.g., docker containers or Kubernetes pods) according to the programmed SLA. The platform, called Network Digital Twin enables the monitoring of any network in real time and can predict reconfiguration actions and potential downtime in advance. The merit of the proposed platform is to be able to provide a virtual representation of the communications requirements, along with its operating environment and real network capabilities. The presentation includes experimental results obtained from several use cases in real industrial settings. Results show that the platform can substantially reduce the testing cost and enable new use cases, like AGV (Autonomous Guided Vehicle) and drone-driven automated network testing as a service.