One of e-Zest Solutions’ client is among the leaders in the Claims Adjudication business, having an experience of more than two decades in the business. The client’s goal was to improve their position in the market, gain an edge over competition and enhance security of business sensitive information.
Given this initiative, e-Zest Solutions was approached to drive digital transformation by helping them overcome consistent business challenges faced in on-boarding new customers from other platforms, as well as provide claims rules alignment, workflow automation and on-demand data extraction for reporting.
The proposed solution was to address efficient execution of the Claim Adjudication process, including: Seamless on-boarding of new customers from one platform to another, irrespective of varied rules on the platforms; detection of defects and anomalies in the claims; identification of mismatches between platforms rules as well as their alignment.
The approach to solve the problem described above, especially with such a massive dataset, required comprehensive use of all the capabilities of data engineering.
Records from one adjudication platform had to be compared with those of another platform, wherein big data was leveraged to process millions of records daily and yield the results of mismatched records.
The data of mismatched records was further utilised using data science to detect various anomalies, defects and exceptions that are violating the claims reimbursement rules across the two adjudication platforms.
The data in numbers, tables and logs was not only massive, but also difficult to analyse and gain insights from. To tackle this, e-Zest Solutions’ Tableau data visualization experts developed a variety of dashboards showing different graphs of clustering groups, from Bubble, Pie to the more conventional but intuitive Column-Stack, among other useful graphs. This enabled the client to gain crucial business insights and make useful decisions to prevent any potential claims errors.
e-Zest Solutions’ chief focus has always been to drive innovation, especially through automation and transformation of existing processes using advanced technologies. With this very aim in mind, e-Zest Solutions brought in innovation for the client by leveraging the following in its solution:
Defect prediction and clustering: This involves grouping of similar cases of claims and then assigning them to make informed and accurate underwriting decisions. The algorithm predicts the defect before it occurs and raises an alarm for underwriters to investigate a case.
Decision tree: All the historical data contained in legacy files of claims certification was used to create a decision tree, which helped in avoiding incorrect decisions and accurate validation of claims.
Anomaly detection: The machine learning algorithm helped detect anomalies.
Post the deployment of the solution, the use of AI has helped create a leaner QA process, with 60 per cent cost reduction and improved accuracy.
Due to automation of a number of manual tasks, not only has the testing and QA team’s size been reduced, but has resulted in improved effectiveness and accuracy too. The automation, as well as anomaly detection and prediction, has helped reduce risk of redundancies, duplication and errors.
If you have an interesting article / experience / case study to share, please get in touch with us at email@example.com