On Comcast’s road to 10G, the virtual cable modem termination system (vCMTS) platform plays an integral part in ensuring the latest Data Over Cable Service Interface Specification (DOCSIS) technology is available to offer higher speeds and the best service to customers. The vCMTS platform is expanding rapidly, which enables enhanced flexibility, scalability, and cost-effectiveness. Therefore, ensuring the stability of this platform is critical, given its significance. However, the introduction of software upgrades often poses challenges for network operators, as it requires meticulous monitoring to detect and address any potential anomalies that may arise in the post-upgrade phase. To overcome these challenges, several systems have been put in place, such as the Automated Network Health Checks. These systems monitor the network's health immediately after deployments or software upgrades, and tools are available for alerting based on a set of key performance indicators (KPIs). While these instantaneous checks are valuable, certain KPIs and telemetry metrics may take hours to days to reveal underlying service degradation. At present, there is limited visibility into detecting these anomalous trends on Distributed Access Architecture (DAA), which could indicate a slow degradation of the platform's health. This paper proposes a comprehensive solution that leverages data science, machine learning, and big data techniques to continuously monitor and detect anomalous trends that could indicate a slow deterioration of platform health and then alert the operations team of any unusual activity.