Accelerating Time to Value through Collaboration

In today’s world of expanding IoT sensors, devices and networks, there is no shortage of information on which to base decisions. OEMs are striving to make the most of this enhanced information for two major initiative areas. First connected shop floors to drive internal operational efficiencies, and second connected products to help end-clients maximize profitability when utilizing the product. For both branches, common elements from edge to cloud are needed to drive solutions, and diverse ecosystems are proving to be effective in accelerating progress. In this context it is important to consider the ecosystem participants and understand how they complement one another. Most projects follow the same basic four-phase approach:

  1. assessing the potential
  2. small-scale proof of concept
  3. industrializing
  4. scaling

Having the right players and technologies connected in the ecosystem is key, and the remainder of this article will focus on how the right ecosystem can accelerate the four-phase approach. First, in assessing the potential, this should always be done with participation of all the key players of the ecosystem that will be involved in providing the solution as most importantly include one or more end-client early adopters. Having this joint participation leads to the most complete value engineering assessment of the benefits that can be generated with the proposed project. Next, having a flexible IoT platform that can be utilized to quickly implement small scale proof of concepts at low incremental cost is important for two reasons. Initially, this IoT platform speeds the lay-down of the proof of concept, and longer term it accelerates progression into industrializing and scaling. With SAS expertise lying in analytics, partnering with an IoT platform powerhouse in Siemens is helping to enable acceleration of the fourphase pattern, and we will use one example to help illustrate this.

Reducing Turbofan engine’s unplanned downtime

The OEM in this example had been looking for ways to meet its clients’ need for greater uptime. The OEM saw AI and IoT as a way to provide applications to help monitor, service, and analyze equipment operations and performance for their customers. While in-house attempts to develop such capability were well received, it became evident for the OEM that developing an industrialized, scalable, and production-ready solution was a task requiring best of breed solutions and expertise.

The solution required the creation of a real-time analytical application that could provide early detection of the Turbofan performance degradation and alert on performance-related anomalies. The OEM’s clients required the AI application to execute on the edge, within their OT environments. Since the assets produce a tremendous amount of data and require minimal latency to enable a valueadded action to be taken, Siemens Industrial Edge and MindSphere were selected as the operational platforms. The Siemens platforms meet the customer’s requirements and provided a mature, readymade solution that could be tightly integrated with the existing assets and OT systems. The combination of these factors streamlined adoption and accelerated time to value.

SAS provides an integrated application for the Mindsphere cloud and the Siemens Industrial Edge platforms, branded as Siemens Edge Streaming Analytics. To enable acceleration of the four-phase pattern described earlier, the SAS platform was used to explore various asset performance and reliability related algorithms, during which Support Vector Data Description (SVDD) was selected as the best performing model. The model was then validated, registered, deployed, and operationalized from the Mindsphere cloud to edge devices connected to clients’ OT environments, managed by the Siemens Industrial Edge platform on-site. A visual dashboard is provided to the Turbofan operators with KPIs for their respective operations with alarm notifications on performance degradation based on Support-Vector-Data-Description (SVDD) analysis. The operator can then make the necessary adjustments to machine settings or further investigate to troubleshoot the asset and guarantee the asset uptime. Over time, monitoring the performance of the analytic model is critical as asset operations may change, equipment degrades, and business requirements change. If no action is taken from a degraded analytic model, the value realized will also diminish. SAS and Siemens together support the continuous and iterative process of retraining analytics and algorithms to meet the ever-changing business needs and utilize Siemens pre-integrated cloud to edge connectivity to deploy the enhanced models directly to the OEM clients’ OT environments.

The example from this OEM is showcased in more detail in a presentation delivered at SAS Global Forum 2021. The example demonstrates a high-level framework of how an organization can realize value in enabling predictive maintenance with their connected assets, utilizing Siemens’ industry-ready platforms and SAS Analytics which is embedded into the Siemens platforms.

About SAS in IoT
SAS empowers organizations to create and sustain business value from diverse IoT data and initiatives, whether that data is at the edge, in the cloud, or anywhere in between. Our robust, scalable, and open edge-to-cloud analytics platform delivers deep expertise in advanced analytics – including AI, machine learning, deep learning, and streaming analytics – to help customers reduce risk and boost business performance. Learn more about our industry and technology solutions at


Kevin Kalish, Principal Industry Consultant, SAS IoT


Craig Foster, Principal Business Development Executive, SAS IoT