Michelle: What differentiates this “radical collaboration” approach from the collaboration we have seen so far?
Razwan: Traditional collaboration has been invaluable but often operates within siloed boundaries. Radical collaboration is about creating a shared data ecosystem where anonymized data from all operators – integrity assessments, in-line inspections, and even satellite imagery – can be pooled together. This comprehensive dataset would allow AI to detect failure mechanisms and assess risks with a level of accuracy that is currently unachievable. It is a shift from sharing best practices to sharing raw insights that drive collective industry-wide improvements.
Michelle: Are there any examples from other industries that illustrate the power of this approach?
Razwan: Absolutely. Take Airbus’s Skywise platform in aviation or John Deere’s ecosystem in agriculture. These industries have harnessed shared data platforms to enable predictive insights and improve efficiency. In each case, the companies set aside competitive concerns in favor of shared benefits and created systems that drive operational improvements across the board. For the pipeline industry, adopting a similar model could mean safer pipelines, more effective corrosion monitoring, and proactive risk management – all made possible through data collaboration.
Michelle: How does AI specifically benefit from this shared data model in the context of pipeline integrity?
Razwan: AI thrives on data diversity. In the pipeline industry, AI can analyze patterns in historical data and make predictive assessments, but its capabilities are limited by the data each company has on hand. Training AI on a broad, anonymized industry dataset would act almost like a “digital mentor” for the entire sector, providing insights based on collective experience. This approach is invaluable for preventing failures and enhancing maintenance strategies, as AI could learn from a vast array of conditions and historical incidents.
Michelle: You mentioned a “digital mentor.” As you know, I have worked extensively on mentoring programs, which is a topic close to my heart. How does that concept play into the industry’s workforce challenges, particularly with the “silver tsunami” of retiring professionals?
Razwan: The digital mentor concept is crucial. When experienced professionals retire, they take decades of specialized knowledge with them. AI trained on comprehensive industry data can capture some of that expertise. For example, a less experienced engineer could use AI-driven insights to make complex assessments, benefiting from the patterns and knowledge embedded in the data. This ensures continuity and accelerates workforce development, helping to bridge the gap left by the retirement of skilled professionals.
Michelle: If AI, as a “digital mentor,” can help preserve and disseminate the expertise of retiring professionals, how do you see this approach influencing competence development across the industry, particularly for younger professionals?
Razwan: Beyond bridging the knowledge gap, AI-driven insights from a shared data ecosystem can create on-demand learning opportunities tailored to professionals at all levels. For younger engineers, it is akin to having access to a repository of industry knowledge, allowing them to learn while performing real-world tasks with AI-informed guidance. This accelerates skill development and builds confidence in handling complex challenges. Additionally, integrating AI with training programs can standardize competence across the workforce, ensuring consistent performance and enabling the industry to set new benchmarks for excellence. It is not just about retaining competence – it is about continuously elevating it.
Michelle: That sounds transformative. Beyond safety, are there other operational benefits to this shared data ecosystem?
Razwan: Definitely. A shared data ecosystem could allow operators to benchmark their corrosion control, inspections, and maintenance spending against anonymized industry averages. These insights would enable companies to optimize their operational expenditures, making smarter, more cost-effective decisions. Beyond that, a shared ecosystem is the only truly effective way to unlock the full potential of AI-driven analytics. By aggregating data at scale, the ecosystem enables the creation of advanced machine learning models that can identify patterns, trends, and inefficiencies that would be impossible to detect with siloed datasets. The larger and more diverse the dataset, the more robust and predictive the AI becomes, providing insights that drive not just operational improvements but also strategic resilience across the entire industry.
Michelle: Radical collaboration offers immense potential. But what barriers need to be overcome?
Razwan: The main challenges are cultural and technical. Data privacy is a long-standing concern, and companies worry about losing competitive advantages. However, these concerns can be mitigated with the right frameworks and standards for secure data governance – something organizations like ASME and API could help establish. Ensuring data interoperability and standardizing data quality are also essential. When companies know that shared data meets consistent standards, integrating it into AI platforms becomes seamless and valuable.
Michelle: What does the path forward look like? How can the pipeline industry turn this vision into reality?
Razwan: The key is to embrace radical collaboration as a strategic advantage. It is not just about sharing data but about setting a new standard where insights drive smarter decisions and elevate industry-wide resilience. By fostering this data-rich environment, pipeline operators and stakeholders can collectively improve safety, operational efficiency, and energy security. Other industries have shown it is possible, and now it is our turn. The time to act is now, and by working together, we can unlock the full potential of AI and pave a sustainable path for the pipeline industry.
Michelle: Razwan, for pipeline operators interested in adopting this approach, what would be a good first step to get started?
Razwan: A great starting point is to focus on specific, impactful applications of data-driven insights. For example, at ROSEN, we are developing data-driven risk assessments and data mining dashboards to support more informed decision-making when assessing pipelines. These tools allow discipline experts access to vast quantities of data – held in our Integrity Data Warehouse – and provide the capability to identify short- and long-term trends, perform risk prioritization, and support predictive modeling. By equipping their teams with these resources, operators can leverage the full potential of data analytics, enabling more informed decision-making and uncovering opportunities for optimization and resilience. It’s about empowering experts with the insights they need to drive impactful change.
Michelle: Thank you, Razwan. It is inspiring to see a vision that combines technology, collaboration, and expertise to tackle the pipeline industry’s most pressing challenges.