Article Synopsis :
The Accenture Labs report for 2018 highlights some of the innovative work being done in the company.
Perhaps of greatest interest is where the work has been successfully deployed and it breaks it down into a number of key areas.
Taking the pith out of claim forms
Everyone keeps saying that AI is coming, but Accenture can point to examples of where it has delivered a meaningful difference to its client’s business processes.
One example was how it helped to speed up claims processing by successfully building a tool that could extract data from ‘first notification of loss’ (FNOL) forms. This has to go beyond scanning forms to understanding the interaction of the data across the form.
Computer says no problem
The Dublin based Accenture Fit team developed a model capable to recommend a personalised training plan for any employee. This saves a lot of trial and error and inertia. Showcased at the Oracle Summit 2018, it is set for integration into the Accenture Foundation Platform for Oracle.
AI reaches the parts mere humans fail to reach
AML compliance is a considerable drain on resources, not in the least due to the vast number of false positives processes produce.
The Dublin Lab developed proofs of concept that address challenges such as entity resolution – understanding that two parties are actually the same – and link analysis, which establishes relationships between parties.
The value was proved by letting the tool loose on the Panama Papers and Paradise Papers. These datasets are too voluminous for human analysts to handle.
A stickler for quality
Quality control ensures customers get the best product from a line, but adds cost to the process.
Accenture explored how computer vision technologies can support quality control.
Working with an automotive supplier, it created a system that identifies when accessories are missing and examines newly assembled car seats to determine whether rework is required.
This meant dealing with irregular shapes, subtle texture variations, varying seat models and different colours. The solution will now be scaled up to a wide number of potential industrial applications.
Using machine learning to manage compliance obligations
The specific “actionable obligations” placed on each entity by regulations can now be automatically extracted. More recently, a “regulatory radar” has been developed to monitor news sources and identify stories that are relevant to both regulators and the companies they regulate.
This helps both keep up to speed with key developments within compliance and identify potential problems.
The success stories are just within the AI space, but the paper covers digital experiences, cybersecurity, systems and platforms and application engineering.
It also discusses how the labs are working with other parts of the industry, other benefits for society, and where the tech is likely to go next.
Link to Full Article:: click here
Digital Insurer's CommentsIf insurers thought that insurtech was a fad and that the old ways will survive, this paper will take the wind from their sales.
There is a vast amount of information in this paper as to how insurtech – and tech in other sectors – is revolutionising how businesses operate and, ultimately, interact with their customers.
Benefits here are not just about improving processes, but also how it can contribute to making healthcare not only more effective but safer.
Link to Source:: click here