Strategic versus Operational Decisions
Every day, small and large companies alike make strategic and operational decisions that influences the bottom line. Strategic decisions are typically made by the C-suite, and these generally are one-off decisions that are made over time and only after careful study of curated information from several sources and consultation with experts. Examples of strategic decisions include mergers and acquisitions and large capital expenditures.
Operational decisions on the other hand are made every day by workers and operations personnel. For small organizations, this can mean dozens, if not hundreds of decisions every day. Elevate this to large organizations, we are now considering thousands, if not millions of actions that impact day-to-day business operations. An example of an operational decision is operation workers deciding to expedite maintenance on a production machine that is not performing as expected. The impact of this type of operational decision can be quite significant as it requires balancing the risk of a lengthy unscheduled downtime event (due to costly breakdown of a critical asset) versus the financial impact due to loss production output.
Operational decisions regarding assets are extensive across many industries since they all rely on the proper operation of countless assets to achieve business value.
Marketplace Challenge(s) / Opportunity
Complicating operational decisioning are challenges due to the pandemic (shift-to -remote), global supply chain disruptions, changing customer demands and loss of skilled workforce and short supply of talent. The “new normal” that is transforming the world as we know it requires ways to tap into the judgement & experience of subject matter experts and combine that with technologies that help organizations improve decision making processes.
Rapidly maturing technologies such as IoT (connected assets), edge computing, artificial intelligence and low code OT platforms designed for subject matter experts provides a unique opportunity for organizations to address operationalizing decisions at scale and the time is now.
So, how can organizations realize the business value of operationalizing decisioning for assets at scale? Here are four essential building blocks:
1. Enabling the Intelligent Asset: Connecting physical objects through a digital network back to compute at the edge or cloud is essence of Internet of Things. The continuous monitoring of sensor data streams that have been unanalyzed is the new frontier being driven by the connected asset. The opportunity for Intelligent Assets is endless. This includes high value industrial assets such as motors, compressors, and pumps; mobile assets such as trains and trucks; remote assets such as wind turbines and solar panels and even Starbucks coffee machines. The low-cost options for sensing, edge compute and network communication options make it possible to generate value for all these assets, especially wherever there is scale.
2. Automation of Analytic Lifecycle: With automation of AI/ML pipelines (AutoML) becoming more common, there is now the opportunity to provide a low-code / no-code interface for operators to execute an automated approach to build and deploy real-time analytics that are connected to IoT streaming data from Intelligent Assets. Deployment options for edge and/or cloud exist and are best determined on a use case basis (type of inferencing, volume, and velocity of data).
If you consider a solar farm, the efficiency and proper maintenance is subject to variability of several sensors such as current, voltage, temperature, and Irradiance. Being able to detect anomalous behavior versus performance impacted by environmental conditions is greatly aided by operators who have experience. Being able to provide operators with the ability to define the sensors and necessary calculations (feature engineering) can go a long way in ensuring the automated ML pipelines are trained to forecast and predict vital parameters that the operators can then interpret and act upon.
3. Low Code Operational Platform For Decisioning: To build an application layer at speed and scale, organizations need to leverage their business operators in addition to developers. Business users provided with rapid development; no code interfaces increase the accuracy of operational decisions since the insights delivered by IoT based streaming analytics is explainable in a language workers understand. For example, instead of an operator needing to interpret a line graph of sensor values and obscure sensor tag names, the interface will provide the context for an alert that they helped define (e.g., coolant temperature is high and fluid level is low). This context greatly aide’s the business expert’s ability to accurately assess the situation and determine the best operational decision.
4. Continuous Improvement: Volvo Trucks has exemplified a continuous improvement approach since they launched a new remote diagnostic as a service for operators with AIoT that included adding hundreds of sensors on Volvo trucks to gain insights that would drive operational efficiencies. Over time, feedback from operational decisions and performed maintenance have been instrumental in helping refine & increase the value of real-time analytics (increased diagnostic coverage & lower false alerts).
The results have been impressive as Volvo has been able to reduce diagnostic time by 70% while reducing repair times by 25%. Today, this innovative service has scaled to over 400,000 trucks and delivered millions of dollars in increased uptime for customers.
Fig 1. Analytical Lifecycle for Operationalizing Decisioning
The formula is simple (See fig. 1). Data from Intelligent Assets needs analytics to achieve insights, and insights needs operational decisioning (actions) to realize value. And for people and operational systems/processes to make decisions, the insights must be operationalized or put into production. Once in production, this can result in closed-loop problem solving powered by real-time insights and this will enable an organization to have an iterative, fail-fast, learn-fast, agile process that provides timely access to insights, resulting in better, more informed operational decisions, at enterprise-scale.
Join SAS at IoT Slam on June 22nd-23rd to see a demonstration on how SAS’ IoT partnerships are enabling operational decisioning at scale, and read this Brief about our collaboration with ClearBlade to learn more.
By: Paul Venditti
Advisory Industry Consultant, IoT
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 www.sas.com/iotsolutions