It seems that AI has suddenly come into its own almost overnight with natural language models like ChatGPT grabbing global headlines. C-SMART though has been quietly developing Augmented Intelligence for a decade, blending sophisticated algorithms, digital twins of gas measurement stations and their manufacturing and calibration specs, and continuous SCADA monitoring to reduce a typical 20 hour work week for measurement analysis to just 2 hours, all while improving results by a factor of 20.
Now that AI is top of mind (and more importantly, available and ready to deploy), let’s dig deeper into what it means for the gas value chain and why teams should care.
Empowering Gas Measurement Teams with Augmented Intelligence
Augmented Intelligence, also known as intelligence amplification or cognitive augmentation, refers to the use of artificial intelligence (AI) technologies to enhance human intelligence and decision-making capabilities. It aims to create a symbiotic relationship between humans and machines, where AI systems support and augment human capabilities rather than replacing them entirely. When it comes to automating human work with augmented intelligence, certain types of natural gas measurement tasks are particularly well-suited for this approach.
- Data Analysis: Augmented intelligence can assist in analyzing large volumes of data quickly and accurately.
- Repetitive Tasks: Mundane and repetitive tasks can be automated with augmented intelligence.
- Decision Support: Augmented intelligence can provide recommendations and insights.
- Quality Control – AI systems can be trained to identify patterns and detect anomalies, elevating data quality.
Status Quo Measurement Analysis Holds Back LAUF Improvement
Consider an example where a natural gas pipeline company operates 25 measurement stations. These stations generate approximately 37,800 data points per week and throw 1,000 red flags that, once investigated, might boil down to ~10 real measurement errors (or error precursors to address).
All of this data and manual analysis are ideal candidates for AI given the types of work that it is especially well suited to support. Measurement teams, however, have been slow to embrace new technology and approaches.
In our pipeline example, a team that relies on the status quo approach of sifting SCADA data streams, manually investigating lost and unaccounted for (LAUF) errors, and system balancing would likely spend more than 20 hours on analysis of the week’s red flags. As a result, teams lack the bandwidth to prioritize flags and alerts, leading to reactive maintenance of measurement equipment, long resolution times for errors, and an industry average LAUF of 0.40%.
A Better Way Forward for Gas Measurement Teams
Continuing with the pipeline operator example with 25 metering stations, C-SMART’s AI-powered approach delivers measurable value by shrinking the typical analysis time from 20 to just 2 hours per week. The SaaS platform, digital pipelines, and C-SMART services allow teams to switch from reactive, top-down gas measurement management to proactive, bottom-up, and maintenance by exception where the same level of work can be managed with 10 to 15 C-SMART alerts per week.
Importantly, C-SMART quantifies errors and provides actionable insights to help prioritize resources and resolve measurement errors much faster and more efficiently. And by giving measurement teams back days of their week, just imagine what new levels of productivity and value can be achieved in your own organization by redirecting valuable analyst time from manual and repetitive tasks to other projects.
C-SMART delivers measurable value on many more dimensions. The key performance indicators we track include 25.2 Tcf of gas monitored, 2,295 errors identified, and $63,000,000 saved for our clients. That works out to be $62,602 saved per metering station per year. That’s why we say C-SMART pays for itself.
Learn more about how C-SMART is shifting the gas measurement paradigm from top-down, reactive, and scheduled maintenance to bottom-up, proactive, maintenance by exception in this blog.