Skip to Content

How to leverage data for a granular view on operational risk

Adam Meskini
09 Mar 2022
capgemini-invent

The risks that organizations are facing today are numerous and complex in nature – growing competitive pressure, cyber-attacks, civil unrest, pandemics, increasingly sophisticated fraud, strong regulatory scrutiny, the development of disruptive technologies, to name a few.

These are times defined by upheaval and there has never been so much pressure on risk functions to have a more consolidated view, nor has there been so much at stake for Chief Risk Officers.

With the constant emergence of new risks and threats there is significant pressure to manage operational risks with a global strategy and extended field of vision. But on the flipside, there are huge benefits for organizations who can gain a precise, clear, and granular view on the intricate complexities of their own unique instances of operational risk.

AI, data and the cloud allow organizations a 360-degree consolidated view on risk, affording greater transparency and process digitization coupled with a robust framework, can help organizations to better identify risks. This will not only bring a clearer view on risk but also ease the burden of risk assessment on humans.

The operational risk frameworks of organizations often lack integration and are a set of fragmented and disparate activities with a diverging variety of risks. But risks interact in a complex manner with big potential for overlap and cross-contamination. However, a much more comprehensive and holistic view is needed. Data and AI, when applied with the right expertise, enable a more fluid risk identification, a more systematic risk assessment, and several capabilities towards risk anticipation.

In Financial Services

Over the past years, the Financial Services (FS) as a sector has been buffeted by penalties that pertain to the mishandling of customers and market misconduct. Add to this the way increasing data permeation and new business models are revolutionizing the way banks serve customers (a recent report from Capgemini Research Institute revealed that only 34% of banks believe they have the business model innovation that they need),  you have an environment in which operational risk capabilities can suffer..

But FS leaders in the application of AI and ML already see it bearing fruit with measurable impact. Artificial Intelligence and deep learning allow the weak signal analysis of the vast datasets of the banking ecosystem so that they are better able to detect the anomalous behavior of rogue traders. Such fraudulent behavior is a massive risk to banks. Over two decades ago, in an era that came before the advent of AI capabilities, the rogue trader Nick Leeson caused the catastrophic collapse of Barings, London’s oldest merchant bank and banker to the Queen. AI, big data, data analytics provide banks with the means to track, predict and prevent such illicit trading.

In Manufacturing

Although there is no sector in which the proper leveraging of AI and data in operational risk wouldn’t bring significant benefits – being able to foresee or forestall hazards is of benefit to any organization – but in manufacturing, the upsides are particularly easy to discern. There’s no good time for downtime on the shop floor. So, being able to detect new drivers and emerging risks to the production line, with weak signals analysis on supply chain disruption can bring much increased agility in the face of future headwinds.

A shot in the arm for risk assessment

The pandemic brought huge global disruption and suffering but valuable lessons have been learned about risk and the potential of data to anticipate outbreaks. The pandemic showed that many organizations simply do not have the flexibility and views on risk that are granular enough, with considerable industry or geo-specificity. Better risk management systems simply could have helped many organizations respond faster and more effectively. Bluedot, for example, an outbreak intelligence platform, was able to identify the emerging risk of an outbreak of COVID-19 eight full days before the World Health Organization; proving that time and data are very much of the essence. By analyzing public health sources and 10,000 articles per day in mass media, and flight itineraries in real time data showed its capability to shorten risk assessment from days to minutes. 

Our Six Demonstrators

At Capgemini Invent we’ve developed 6 key demonstrators that, by leveraging AI, Data, and the cloud, help organizations gain more complete and granular view on risk they need:

Dynamic RCSA (risk control self-assessment) helps to adapt the risk appreciation of a specific kind to the precise nature of a given organization. It automatically generates an individualized set of risk and control libraries based on real-time auditing, reducing manual and repetitive workload in turn giving real-time RCSA results.

The RCSA Assessment Tool facilitates the RCSA Assessment, simplifying it with a common methodology. It consolidates RCSA assessments to provide a global overview of risks. Benefits include a normalized process and methodology for RCSA assessment with a greater degree of control easing the collection process at each assessment phase.

RCSA Consolidation/ visualization gathers a consolidated view of risks across the organization, using advanced visualization and analytics technologies. It’s a collaborative tool that supports both business and operational teams in their decision-making through a consolidated view by risk area. This eases the identification of priorities regarding control framework improvement and supervision, and garners the information needed for internal and external reports.

Benchmarking of internal /external risks provides the powerful exploration of external incidents to provide insights for increased accuracy in risk assessment through appropriate benchmarks (generic and specific). It brings support risk detection and identification by analyzing relevant databases of historical incidents.

 The Taxonomy Rationalization and Classification Tool automatically classifies operational risk incidents into an internal taxonomy of risks based on incident descriptions. It assists risk managers with advanced exploration and visualization capabilities for better assessment of operational risks. By automating benchmarking of metrics and repetitive and time-consuming tasks, it frees up humans for more roles that add value and drives efficiency.

 The Weak Signals Analysis tool reveals an organization’s most recurring operational risks, materially relevant topics, and related causes, based on incident descriptions. It provides a more powerful lens for the detection and precise definition of Key Risk Indicators (KRIs). It allows for the faster identification of risk drivers and emerging risks based on inputs provided by AI data extraction and analysis.

In conclusion

There are many benefits of leveraging data and AI to gain a clearer and more granular view on risk. Imagine you knew exactly how many mosquitoes have found their way into your bedroom, and exactly where they were, before you went to sleep! You would be able to locate and neutralize the threat and avoid significant annoyance, discomfort, and disruption. Data, AI, and the cloud can provide organizations with such a granular view on risk and bring huge benefits if only they are applied in the right way.

If an organization can lessen and avoid risks, it can become a powerful one indeed, but the laggards in leveraging AI and Data in the best way will remain open to those annoying bites in the dark. If you can see it, you can squash it.

Author

Adam Meskini

Senior Manager, Risk Management Powered by Data, Capgemini Invent
Creating and scaling value for Risk & Corporate functions