With all the buzz, press and enthusiasm surrounding ChatGPT, more and more oil and gas professionals are exploring how ChatGPT and similar AI chatbots can help them improve their work and make them look smarter to their bosses.
Large language models (LLM) enable AI chatbots to evaluate their vast amounts of raw data to respond to end-user input. They are causing enormous debate about the future of many professions, including those in the oil and natural gas industry.
How can ChatGPT help? Examples of oil and gas work products where AI chatbots can help include geological interpretation, seismic interpretation, software development, drilling designs, engineering reports, and legal contracts.
Deepen data analysis
AI chatbots can quickly and accurately analyze large volumes of data from diverse sources. This capability is valuable for oil and gas professionals who must query, process, and interpret large amounts of structured and unstructured data for exploration and production.
AI chatbots overcome the limitations of current data analysis tools, including:
- Poor response times for large queries.
- Support only simple text string matching for unstructured data.
- Complexity in describing queries for structured data.
- Difficulty accessing data stored on various platforms.
- Difficulty integrating data due to incompatible key formats and values.
AI chatbots can deepen data analysis by searching relevant information to identify patterns and insights by:
- Responding to more sophisticated queries, such as those that include geology and production.
- Handling complex data integration more easily, including integrations with industry data providers.
- Analyzing vast amounts of unstructured data such as industry reports, literature and web pages.
- Summarizing text from many regional studies.
Improve diagnostics
AI chatbots can quickly and accurately analyze large volumes of diagnostics data from Industrial Internet of Things (IIoT) sensors at gas processing plants, compressor stations and terminals. This capability is a valuable digital application for oil and gas professionals who analyze production and refinery machinery performance data with a view to:
- Improving compressor performance.
- Predicting failures and reducing unscheduled outages in ESPs.
- Diagnosing root causes more quickly and accurately.
The following issues limit the current diagnostics process:
- The complexity of modelling software required.
- Data volume and data source maximums.
- Insufficient forecasting and data visualization functionality.
AI chatbots can improve diagnostics by identifying the following:
- Patterns and relationships that analytics software misses.
- Targets for intervention earlier and more quickly.
- Optimizing schedules for planned maintenance outages.
Improve forecasting
It’s impossible to predict the future. However, forecasting software can help oil and gas professionals avoid guessing and narrow the confidence interval of future scenarios for product prices, transportation, and operating costs.
AI chatbots can overcome the limitations of current forecasting software. These limitations include the following:
- Simplistic models leading to low-confidence forecasts.
- Restrictions on the amount of data that can be ingested.
- Lapses in data quality casting doubt on the usefulness of forecasts.
AI chatbots can improve forecasting by:
- Using more sophisticated and likely more realistic price forecasting models.
- Considering more data sources and more extended time series.
- Including data sources, often external to the organization, the forecaster did not consider.
Write reports and software
Writing reports, presentations, and software has always been a hand-crafted and frustrating creative process for oil and gas professionals.
A rapid digital transformation of these tasks occurred for many professionals with the introduction of:
- Word processors and text editors for reports.
- Sophisticated editors for graphics, photos, audio and video.
- Integrated development environments (IDE) for software.
AI chatbots are now taking this digital creative work to a new level by generating reports, designs, software, graphics and video based on a surprisingly small amount of text input about what is wanted.
While AI chatbot products still require tuning and correction by professionals, AI chatbots have significantly reduced the associated effort. Three early examples are:
- Software development: A GitHub research project reports that using the GitHub copilot reduced developer effort by approximately 50 percent and improved developer satisfaction.
- Photography: Using the text-to-graphic feature of some AI chatbots can reduce photographer effort from hours to minutes by avoiding travel time.
- Video monitoring: Some AI chatbots can review reams of video from facility surveillance to identify intruders and potentially catastrophic incidents.
Manage knowledge
Oil and gas professionals recognize the value of knowledge management to more rigorously apply the experience of others and from one project to the next. They also understand the impediments, including:
- Staff turnover.
- Gaps in the documentation of previous exploration and production projects.
- Changing technology.
- The cost of operating a knowledge management system.
- Evolving customer and regulatory requirements.
Some organizations have introduced software packages to knowledge management by:
- Advancing digital transformation in their work to capture the requisite digital data.
- Operating centres of excellence for various professional groups.
AI chatbots can advance knowledge management by:
- Making internal data easily accessible and searchable.
- Providing excellent references to external sources for most topics.
- Generating new literature on most topics, including combinations of hypotheses and theories that could guide future research.
ChatGPT has triggered a debate about the future of knowledge management. Some believe that AI chatbots will finally cause knowledge management to be taken more seriously and adopted more widely. Others believe AI chatbots will eliminate the need for formal knowledge management approaches and principles because AI chatbots will make so much knowledge readily accessible.
Reduce mind-numbing administrative overload
A surprising number of organizations continue to handle many routine administrative transactions manually. Oil and gas professionals are invariably sucked into error resolution in systems including accounts payable, regulatory applications and records retrieval. Some organizations have implemented robotic process automation (RPA) to:
- Improve the work experience of their employees.
- Accelerate processing to increase efficiency.
- Increase processing accuracy.
- Reduce the number of exception situations and errors.
Unfortunately, current transaction processing software is constrained by the following:
- Customer and vendor errors.
- Limits on the sophistication of automated processing rules.
- Errors interpreting scanned documents.
AI chatbots can improve transaction processing by:
- Handling complex processing rules.
- Reducing the error rate.
- Increasing throughput.
AI chatbots' capabilities in NLP, knowledge management, data analysis, diagnostics, and automation make them valuable assets for oil and gas professionals looking to enhance their work.
Yogi Schulz has over 40 years of information technology experience in various industries. Yogi works extensively in the petroleum industry. He manages projects that arise from changes in business requirements, the need to leverage technology opportunities, and mergers. His specialties include IT strategy, web strategy and project management.
