Mastering the Art of Unsupervised Learning: Techniques and Guidance
Unsupervised learning, a cornerstone of artificial intelligence, has emerged as a powerful technique that enables machines to uncover patterns and insights from unlabelled data.
Big data is not new to the oil and gas industry. Since 1980s digital technologies has been integrated to understand reservoir’s resource and production potential. However, there has been a large gap after then and how the industry is not making the most of what technology, data, and AI can do for the oil and gas industry.
There has been selective adoption of technology and unsystematic implementation of digitalization. To date, with more pressure on finding limited resources, there is much more potential and reason to integrate AI and ML into the operation to boost operational efficiencies, improve health and safety, and increase business value.
Oil and gas sector at large has invested proportionally less on digital and AI/ML-related technologies in the last decade compared with other sectors, such as banking, automotive, health care, retail and consumer products, and software. This is partly because it’s seen as risky, unproven technology and it requires highly skilled programmers and data scientists. But it’s also because it requires a sustained, long-term investment of dollars that many companies simply couldn’t afford during the downturn.
In addition, digital fluency and the use of AI/ML technologies have not been seen as core competencies in a sector that’s long been ruled by human operators — namely, mechanical, chemical and electrical engineers. Early adopters are typically sectors that are comfortable with digital fluency and those that have access to standardized data sets.
But the time is changing. According to a paper by with World Economic Forum various macroeconomic, industry and technology trend is reshaping the industry now and it’s looking into a new wave of digital and business technologies. This is propelled by:
Disruption in supply, demand, and commodity prices
The industry had its worst downturns driven by supply-side disruption with commodity prices falling to more than 70%. Just as signs of recovery emerged, another disruption is imminent due to the peak demand for oil. These disruptions will maintain pressure on hydrocarbon prices and prompt energy companies to focus on reforming their portfolio in the energy transition.
Rapid Advances in Technology
Oil and gas companies now have the capacity for real-time decision making and execution and enhanced agility due to the growing sophistication of technology platforms, mobility surveillance, connectivity, storage techniques and the ability to process and analyse data rapidly.
Changing consumer needs and expectations
Increased consumer expectations on personalisation, engagement, speed, and sustainability are driving their energy choices. Companies are now influenced to be more transparent in their operation including hydrocarbon sources and emissions and to be more connected to multiple technology and digital platforms.
There are four themes central to the digital transformation of the oil and gas industry in the next 10 years.
Digital Asset Life Cycle Management
New digital technologies combined with data-driven insights can transform operations, boosting agility and strategic decision-making, and resulting in new business models.
Circular Collaborative Ecosystem
Applying integrated digital platforms enhances collaboration among ecosystem participants, helping to fast-track innovation, reduce costs and provide operational transparency.
Beyond the Barrel
Innovative customer engagement models offer flexibility and a personalized experience, opening up new revenue opportunities for Oil and Gas operators, and new services for customers.
Energizing New Energies.
The digitalization of energy systems promotes new energy sources and carriers, and supports innovative models for optimizing and marketing energy. To remain relevant to customers, the Oil and Gas industry must understand the full impact of these changes on the broader energy system.
Digital transformation in the Oil and Gas industry could unlock approximately $1.6 trillion of value for the industry, its customers, and wider society.
It could create benefits worth about $640 billion for wider society. This includes approximately $170 billion of savings for customers, roughly $10 billion of productivity improvements, $30 billion from reducing water usage and $430 billion from lowering emissions.
Environmental benefits include reducing CO2 emissions by approximately 1,300 million tonnes, saving about 800 million gallons of water, and avoiding oil spills equivalent to about 230,000 barrels of oil.
To unlock the values Data and AI in the oil and gas industry, digital transformation must be initiated from the top and embraced throughout the organisation. The World Economic Forum has six recommendations to achieve this.
Make digital a priority for senior executives.
Digital transformation, like any other transformation, needs to be sponsored from the top. This includes setting a clear vision, committing funding and resources, and actively championing the change-management effort associated with it.
Drive a culture of innovation and technology adoption.
While not everything will be developed inhouse, companies will need to open up to new ideas and ways of working.
Invest in human capital and development programs.
Ultimately, a digital-savvy workforce is both a foundational enabler of transformation and a key driver for maximizing value capture.
Put in place a methodical approach for developing and/or industrializing new capabilities.
This includes decisions about whether to build or buy capabilities, and a programmed management approach to scale up the technology and digital platforms.
Reform the company’s data architecture.
Data sits at the heart of digital transformation, so the harmonization, integration and interoperability of data platforms are critical.
Identify opportunities to deepen collaboration and understanding of sharing-economy platforms.
This will allow for sidestepping the potential pitfalls brought by changing customer preferences shaped by the rise of the sharing economy.
Adoption of AI and ML needs to be anchored on business strategy and according to an EY study, in the context of oil and gas, the critical drivers will be:
Cost and efficiency needs to be the driving forces of transformative change programs. There should be a solid business case for AI and ML strategic success.
– How AI and ML integrate into the overall business excellence and cost reduction
– What will be the cost-related business case for change in applying AI/ML?
– What level of cost reduction is required and where in the business do the opportunities lie?
Positioning AI/ML to enable business achieve quality-based decisions that drive deeper insights and value is critical.
– Have you identified the problem the organisation is trying to solve
– Have you applied AI/ML in your business and are they applied successfully to provide quality outcomes?
– Is AI/ML positioned as a tool to drive quality?
To drive adoption, strategy must harness humans and machines to work together collectively to support scaling and creative new capabilities.
– Are there opportunities for humans and machines to work together in a new and different way?
– Can people trust technology tools to help make better decisions and add value?
– What new business capabilities is the organisation missing that it wishes to have?
Human Resources (Hire to Retire)
During boom cycles, AI/ML is crucial for talent management. Using algorithms help HR scan resumes and look for clues with regards to candidates who may be worth proceeding. It can also be developed to match candidates with divisions that best fits their experience and career interests. It can also be applied to workforce management and talent acquisition. Capturing institutional knowledge of seasoned workers who are close to retirement and predicting at-risk high-value employees or talent shortages.
Finance (Cost Allocation)
Complex accounting is involved in the industry and there are a number of opportunities for AI/ML adoption in this area including automating the cost allocation process and cost accounting.
Cost allocation is very rule-based and ripe for AI/ML that could be programmed and taught to learn how to do cost allocations providing opportunities to redeploy specialized accountants to more value-added analysis.
Maintenance (truck engine maintenance planning and execution)
Truck engines and other equipment used in oil sands capture valuable data such as vibration, temperature, pressure, and throughput but these data mostly never gets used. AI and ML can be used to predict engine failure and potentially reduce expensive maintenance issues and increase equipment uptime. With a data lake and ML find out correlations that could drive more predictive insights and create better engine performance and reduce cost.
Capital Planning (Portfolio management)
Consolidation within the sector is a common strategy, large companies acquiring assets and companies in distress. Executives are continuously evaluating their portfolios making high-stakes decisions on what to buy or sell for optimum portfolio performance. AI/ML could help executives evaluate metrics and identify gaps and opportunities to make buy or sell recommendations.
Subsurface (well data analysis)
AI/ML can be used in tandem with humans to process data and analyse complex geological rock formations to assess where hydrocarbons can be found and potential and probable volume. The platform can also be used to improve ways of finding correlations and develop better recommendations on whether to develop and explore further or walk away from, improving investment value.
Environment, Health, and Safety (safety assessment and root cause analysis)
Due to the nature of the business, safety is a primary concern and is a critical indicator of day-to-day operational success. In fact, most companies would have deployed safety management systems and behavior-driven activities and embedded safety leadership in operations. While there is immense data captured on why safety issues occur, their root causes, and investigation to try and prevent future incidents, they still happen. AI/ML has the potential to assess data and learn from it over time to drive deeper insight into root causes and create better preventative decision-making.
Apart from the use of robotics to automate warehouse picking and packing processes, AI/ML has the potential to strategically predict demand, making stocking types and levels a source of value. By using mathematics and predictive analytics, machines can assess historical consumption data, draw correlations and then make recommendations to human warehousing staff, or integrate directly into the ERP system, to help optimize stocking levels, replenishment and network planning between warehouses.
AI Consulting Group can help oil and gas enterprises who wish to adopt data and AI-enabled solutions for their business. Speak to our management consultant to explore the best use cases to help your organization solve your most pressing challenges and achieve quick wins.