Communications on Applied Nonlinear Analysis ISSN: 1074-133X Vol 29 No. 4 (2022) 14 https://internationalpubls.com Ml-Powered Claims Triage: Using Uipath Ai Centre for Intelligent Work Distribution Praneetha Kotla kotlapraneetha@gmail.com Article History: Received: 12-09-2022 Revised: 15-10-2022 Accepted: 01-11-2022 Abstract: UiPath AI Centre allows the system to find out which claims are to be processed automatically and which require immediate human intervention to optimise the work distribution and the claims processing. This research aims to investigate the role of ML-enabled claims with the help of “UiPath AI Centre” to initiate automation, fasten, and improve the accuracy and efficacy of the overall claim process and to achieve this aim, both qualitative and quantitative method was incorporated. By leveraging the AI models, human agents were able to do smart triaging, thereby reducing the need for them to attend to complexities and improving the process of claims resolution for their entire network more efficiently and quickly. It is worth mentioning that such a strategic use of UiPath AI Centre allowed for receiving higher efficiency, lower costs, higher customer satisfaction, and others. Key Words: UiPath, AI, ML, RPA, Insurance sector I. INTRODUCTION A. Background to the Study ML-enabled claims triage through “UiPath AI Centre” indicates leveraging machine learning frameworks within UiPath's platform to automate the screening and categorization of insurance claims. AI triaging for telehealth incorporates “conversational agents” embedded in a virtual or phone visit to stratify patients [1]. It is a period of rapidly changing technology and high expectations for customers, and therefore, the insurance industry is going through a digital transformation. Figure 1: Aim of Transformation mailto:kotlapraneetha@gmail.com Communications on Applied Nonlinear Analysis ISSN: 1074-133X Vol 29 No. 4 (2022) 15 https://internationalpubls.com Transformation is about improving customer experience, improving operations, decreasing cost, and leveraging data in risk assessment, product personalisation and others referred to in Figure 1. A lot of changes call for the adoption of AI, machine learning and automation, the development of digital channels and platforms. However, AI and RPA applied to claims processing in RPA Operational Excellence can assist insurers by managing a high volume of claims, making these claims more accurate, reducing fraud, and settling claims fast [2]. B. Overview ML, or machine learning, enables triage and cultivates AI as well as “UiPath’s AI Centre” to effectively value and divide the claims of insurance. Thus, by incorporating ML along with robotics technology such as RPA, or “robotic process automation,” insurers initiate automation in the informed decision-making process, fasten the workforce, and improve satisfaction of customers and claim assessment [3]. These have been helping to reform the claim management system into a more effective and data-based process in the insurance sector. C. Problem Statement Processing a large number of claims is a difficult operation for insurance companies since they do not want to waste time and resources on mundane tasks while also trying to avoid losing money on paid claims. The delays and the higher operational costs not only put the customer in the wrong, but it is a bad influence on the triaging process itself. These issues are tackled in this paper by looking into how one can use machine learning to automate and improve claims classification and prioritization through the UiPath AI Centre. RPA platforms are generated in on-premise, cloud, and hybrid environments [4]. This explains how AI-based triage makes the system more accurate, quicker, and more productive in that it allows better awareness of workload distribution. D. Objectives The primary goals of this study are: 1. To highlight the role of UiPath AI Centre for Intelligent Work Distribution in the insurance sector. 2. To initiate automation and value insurance claims through ML-based models with “UiPath AI Centre.” 3. To decrease processing time and manual intervention with the help of “intelligent work distribution”. 4. To propose strategies to train reliable ML frameworks. Thus, these research objectives aim to investigate the role of ML-enabled claims with the help of “UiPath AI Centre” to initiate automation, accelerate, and improve the accuracy and efficacy of the overall claim process in the insurance sector. E. Scope and Significance This paper focuses on a use case of using the machine learning trained applications in UiPath AI Centre within the insurance domain. The market share of UiPath was around 13% in 2018, identified as a leading vendor [5]. The study discusses the automation of classifying, prioritizing, and intelligently distributing tasks to improve operational efficiency. Combining AI and RPA with insurers not only enables them to make data-driven decisions, distribute resources effectively, Communications on Applied Nonlinear Analysis ISSN: 1074-133X Vol 29 No. 4 (2022) 16 https://internationalpubls.com and cope with increasing claim volumes, but is a purpose that also enables the insurer to achieve a competitive edge within an increasingly more digital and customer-centric industry. In this regard, there is significance in the potential to transform a more traditional claims management scenario by cutting processing time, reducing errors, and returning better customer satisfaction. II. LITERATURE REVIEW A. Role of UiPath AI Centre for Intelligent Work Distribution With UiPath AI Centre, the company can orchestrate all the moving parts of artificial intelligence and conveniently deploy the limitless power of AI to any software that is being automated today by RPA. “UiPath AI Centre” reduces manual workload, speeds up the processing, thereby strengthening the power of decision making. AI needs to have a coherent initiative to collect data and storage in insurance companies [6]. Figure 2: Stages of information value loop [6] The stages of the information value loop include act, analyse, communicate and others in the above figure. The utility of this includes allowing insurers to deploy, manage, and retrain machine learning models to classify, prioritize, and route the claims by urgency, complexity, and risk. For example, AXA, which is a leading global insurance company, integrated “UiPath’s AI and RPA equipment” to speed up claims handling. Thus, by using AI Centre, AXA automated repetitive tasks, reduced claim settlement time, and improved customer satisfaction [7]. B. Automate the classification and prioritization of insurance claims “The UiPath AI Centre” keeps AI models seamlessly integrated into RPA workflows. Hence, complex or high-risk claims are flagged for review, while simple or low-risk claims are automatically processed. The response times are faster, resource allocation is optimized, and Communications on Applied Nonlinear Analysis ISSN: 1074-133X Vol 29 No. 4 (2022) 17 https://internationalpubls.com customer experience is enhanced. Manual, time-consuming, and open to human error, claims processing used to be a laborious task that would lead to dissatisfied customers and long delays. UiPath acts as an RPA tool used in companies and projects [8]. This way, the automation of insurance claim classification and prioritisation through UiPath AI Centre automation of machine learning models built on structured and unstructured data such as customer information, claim type, and incident details to automatically categorize and prioritize claims according to urgency, complexity, and possible fraud makes it possible for the bulk of a lot of claims to be handled within the insurance company. For example, Zurich Insurance has deployed UiPath’s automation solutions to rewire the “back office” portion of its claims process. As an outcome of integrating the AI into RPA, Zurich was able to automate document processing, have more accuracy with claims, and reduce handling time [9]. Placing the burden on them thus helped the teams of machine learning models to prioritize claims more effectively so that they could focus on complex cases. C. Decrease processing time and manual intervention Intelligent work distribution of claims within the realm of automating repetitive tasks and automatically routing claims based on complexity and priority, among others, significantly decreases the amount of manual intervention and the processing time involved. As a result, fast claims settlement, accuracy, and cost saving are gained. RPA can be used to do things like data entry, document processing, and validation of claims out of the hands of human resources. AI, cloud computing led to use cases for the insurance sector [10]. Many of those claims can be automatically assessed by AI-powered systems based on pre-defined rules and without the need for human review. Automatic document extraction and validation from documents can be done using IDP systems that automatically extract and validate data on documents, reducing the time spent on data entry and verification. Automation subsequently increased the processing speed as it enormously shortened the processing time of claims, thereby reducing the settlement time of claims. This, in turn, allows insurers to automatically handle a higher number of claims in less time. Automating routine tasks allows the organization to devote employees to the finer tasks and higher-level strategic initiatives. By factoring things like urgency, risk, and customer satisfaction in terms of AI priorities, critical claims will be at the top of the list. For instance, Allianz Insurance used UiPath to automate over 50% of its claims-related tasks [11]. It also increased accuracy and reduced the amount of work done by employees. As the responsibility for work distribution is smartly assigned to different units, Allianz streamlined the working pace and increased performance as a whole for its policyholders. D. Strategies to train reliable ML frameworks In case of insurance claims triage using machine learning models, effective data preprocessing is required for reliable models. Normalizing the features helps in scaling numerical data uniformly Communications on Applied Nonlinear Analysis ISSN: 1074-133X Vol 29 No. 4 (2022) 18 https://internationalpubls.com so that no single feature influences the model since it scales non uniformly. “Natural Language Processing” (NLP) is used for structured data like claim descriptions or emails to extract meaningful features using tokenization, stop word removal, named entity recognition, and others. Additionally, there are UiPath’s data labelling tools that help make the annotation of datasets accurate for supervised learning models. The “Garbage In, Garbage Out” (GIGO) theory states that poor quality of input data will produce unreliable outputs to illustrate the significance of good preprocessing for accurate AI predictions and blockchain [12]. III. METHODOLOGY A. Research Design Research design acts as a structured model that navigates the entire plan to execute a project. This paper has selected an “explanatory research design” to explore the role of “ML-Powered Claims Triage.” This research design is integrated as it endeavours to investigate and describe the causal connection between machine learning integration, automation, and claims management efficiency. The explanatory design is a “two-stage” system that includes quantitative data being used as a base on which to build and cultivate qualitative data [13]. This would prove useful to identify the domain meaning of how intelligent work distribution influences claim classification, processing time, and accuracy. Descriptive and exploratory research designs were incompatible because, in these forms of research, it is focused on watching to see what happens without looking into how things behave and what causes automation and efficiency improvement in claims processing. B. Data Collection This research has applied a “multi-research strategy” with the selection of both secondary quantitative and qualitative data collection and analysis techniques. Data sources used for the secondary qualitative research are academic journal articles, case study examples, and industry reports. Furthermore, statistical graphs, charts, and metrics are collected and further interpreted in a secondary quantitative method. This helps to create objective evidence, improving the reliability and validity of the outcomes and enabling informed decision-making. C. Case Studies/Examples Case Study 1: Claims Automation with AI and RPA UiPath and its AI and “robotic process automation” (RPA) turned into a global leader in insurance and automated AXA’s claims processing workflow into an automated one. With the use of machine learning algorithms, AXA was able to make this process automated such that claims could be classified and prioritized automatically and thereby reduce the level of manual intervention as well as reduce response times [14]. Case Study 2: Document Processing and Intelligent Claims Handling Communications on Applied Nonlinear Analysis ISSN: 1074-133X Vol 29 No. 4 (2022) 19 https://internationalpubls.com Zurich Insurance utilised UiPath’s automation solutions to augment the claims processing. Zurich integrated its machine learning models with RPA to automate the document classification, data extraction, and fraud detection tasks so that claim prioritisation is done faster [15]. Case Study 3: AI and RPA for Claims Triage and Fraud Detection Europe’s largest insurance group, Generali, integrating AI and RPA into its claims management system to help enhance claim triage and fraud detection. The claims were categorised based on machine learning models, and ones with higher risk faults were valued [16]. D. Evaluation Metrics Figure 3: Evaluation matrices [Self-Created] Evaluation matrices such as processing time for claim purposes, Accuracy Rate, Customer Satisfaction, and others are proactive to cultivate integration and success of ML-enabled claims triage as well as distribution of the intelligent work. “Customer Value Score” has a significant effect on customer satisfaction parameters [17]. These metrics can measure credibility, performance, and the experience of customers by creating a clear evaluation of the effect of automation technology on performance claim management while guiding insurers in upgrading their methods for optimal outcomes. IV. RESULTS A. Data Presentation Traditional methods are often time-consuming and prone to errors. Solving the problem of classifying and prioritising insurance claims is done by automating the process using ML-powered claims triage integrated with the UiPath AI Centre. In this approach, machine learning models are used by the approach to analyse historical claims data, discover patterns, and make intelligent decisions regarding the handling of the claims, either of immediate action, further verification, or automated processing. Communications on Applied Nonlinear Analysis ISSN: 1074-133X Vol 29 No. 4 (2022) 20 https://internationalpubls.com Figure 4: Distribution of Anomaly Scores in Financial Transactions [18] The above figure has highlighted how anomaly scores vary for a wide set of financial transactions. The majority of transactions report anomaly scores that are close to -0.4, suggesting they are standard [18]. The number of transactions increases dramatically as credit scores get close to this range. A small number of IoT transactions have anomaly scores below -0.6 which could be a sign of possible fraud [18]. The uneven distribution of data helps determine a red flag to pick out unusual transactions for computers to watch out for. Figure 5: Technological Innovation among Insurers [19] RPA allows companies to excel in the insurance market, and our Andersen’s FinTech experts will explain how. The investments of insurance businesses in emerging technology in the last five years have grown by 650% [19]. For 29% of tasks in financial services, RPA is the method used. Communications on Applied Nonlinear Analysis ISSN: 1074-133X Vol 29 No. 4 (2022) 21 https://internationalpubls.com Figure 6: UiPath Deployment [20] As per the above image, UiPath is growing 65% y/y at $580m ARR after 107% [20]. This is increasing effectively and crossed $580m in revenue in the initial years [20]. B. Findings Many transactions are categorised as normal, based on their anomaly scores in Figure 4. A very limited number have scores that indicate there may be fraud. This way, it becomes possible to establish limits for detecting unusual financial actions. [18]. Other factors identified as influencing the growth of the market are the advancement in the technological aspect of IoT and BYOD, along with increasing demand for cloud-based infrastructure. Under RPA in insurance, to use of software robots to automate routine industry-specific tasks. This is often stated in the audience that UiPath is one of the fastest-growing SaaS companies of all time, from $1m in revenue to $600m in revenue within 5 years [20]. C. Case Study Outcomes Case Study Name Company Outcome Relevance to the Paper Claims Automation with AI and RPA AXA Decreased processing charges and time for the statement, improving the satisfaction rate of customers. This case study example demonstrates AI and RPA potential in automating insurance workflows for global Communications on Applied Nonlinear Analysis ISSN: 1074-133X Vol 29 No. 4 (2022) 22 https://internationalpubls.com insurers, accelerating and enhancing the process of claims management. Document Processing and Intelligent Claims Handling Zurich Decreased manual intermediation, streamlined the time to response, and developed the fraud detection technique. This highlights the potential of document automation and the intelligent handling of words to improve operational efficiency in large-scale insurance companies. AI and RPA for Claims Triage and Fraud Detection Generali Group 20 to 255 fasten the time for claim processing and a 10 to 15% decrease in losses related to fraud activities. This case study example shows how AI and RPA are very useful to boost fraud detection and streamline claims processing in the insurance industry, and give an advantage. Table 1: Case Study Analysis [Self-Created] These case study examples show the impact of UiPath's RPA and AI integration to reform insurance claims, credibility and satisfaction rate of customers in insurance companies. D. Comparative Analysis Author Aim Findings Gap [6] This article aims to create and identify a technological environment to detect fraud activities. “IoT to complement other advanced technology-led futuristic implementations.” Lack of statistical analysis [8] This aims to “provide a systematic study of RPA with an example “Robotic Process Automation there is a wide ability for the acceleration in the adoption of Lack of primary research. Communications on Applied Nonlinear Analysis ISSN: 1074-133X Vol 29 No. 4 (2022) 23 https://internationalpubls.com of its use in a sector.” “Robotic Process. Automation”. [10] This article aims to identify the effect of digitalisation in insurance companies. “The digital transformation is of increasing relevance for insurance companies’ business frameworks”. Lack of theoretical incorporation [12] This article aims to identify the impact of digitalisation on global business “Blockchain initiates profound and dramatic effects of digital transformation on the global economy”. Lack of statistical research Table 2: Comparative Analysis [Self-Created] Interpretation in the above table helps to fulfil research aims and objectives by identifying the aim, findings and gap, specifying refined knowledge of UiPath and technology advancement in the insurance sector. V. DISCUSSION A. Interpretation of Results Both secondary qualitative and quantitative analyses were conducted to fulfil the research aim and objectives. Qualitative analysis by highlighting Zurich, AXA, and Generali Group was done to identify the “role of UiPath AI Centre for Intelligent Work Distribution in the insurance sector.” Additionally, graphs have highlighted the market overview, technological trends of UiPath, AI, and ML in the insurance sector to fulfil the parameters of ML-based models with “UiPath AI Centre” [19]. Furthermore, “effective data preprocessing,” “NLP integration,” and others were identified as strategies to improve the accuracy and efficacy of the overall claim process in the insurance sector. B. Practical Implications The impact of “ML-powered claims triage” using “UiPath AI Centre” is relevant and practical for the insurance sector. It makes it possible for insurers to automate repeat tasks, lessen human errors of humans and enable claims processes to be sped up, hence saving costs and enhancing the satisfaction of customers served. Insurers can improve very intelligently how they allocate their resources between work that is more complex and more urgent [4]. This also improves the risk management capabilities of fraud detection. These advancements augment the agile and scalable Communications on Applied Nonlinear Analysis ISSN: 1074-133X Vol 29 No. 4 (2022) 24 https://internationalpubls.com claims management system that enables insurers to get through increasing amounts of volume with particularity in obvious demand or crisis states. C. Challenges and Limitations However, this paper had its limitations, such as too much dependence on secondary data collection and analysis processes that led to a threat of bias in the outcome of this paper. Qualitative analysis was done by showing 2 to 3 instances of case studies, limiting the wider scope of this research [16]. Additionally, there are issues related to data quality and integration complexity between AI models and legacy systems and insufficient availability of labelled training data that limit the success of the research. D. Recommendations Thus, to develop ML “claims triage efficacy”, insurance providers need to invest in a decent quality claims dataset collection and continuous model learning using various claim datasets. The company can further improve accuracy and automation efficiency using UiPath AI Centre’s data labelling and monitoring tools. Integration with legacy systems should take place seamlessly and with data privacy regulations preserved [21]. Domain knowledge is key when fine-tuning AI models based on data science knowledge, so the collaboration of data scientists and claims experts is essential. VI. CONCLUSION AND FUTURE WORK In conclusion, it is proven that the integration of ML and the UiPath AI Centre for quality intelligent claims triage in insurance not only is beneficial but has already seen success. Challenges such as data quality, model training, and system integration were highlighted. Its practical success at leading insurers has been confirmed in real-life case studies. Automation makes operations more efficient, accurate, and customer satisfactory by decreasing manual workload and scheduling resources optimally. Future work needs to be done on improving AI model adaptability, strengthening the data privacy measures, and real-time analytics on dynamic claims processing. Further validating the scalability and long-term impact of the approach could be done on one or more of the insurance domains. VII. REFERENCE LIST [1] Aggarwal, N., Ahmed, M., Basu, S., Curtin, J.J., Evans, B.J., Matheny, M.E., Nundy, S., Sendak, M.P., Shachar, C., Shah, R.U. and Thadaney-Israni, S., 2020. Advancing artificial intelligence in health settings outside the hospital and clinic. NAM perspectives, 2020, pp.10- 31478. [2] Lamberton, C., Brigo, D. and Hoy, D., 2017. Impact of Robotics, RPA and AI on the insurance industry: challenges and opportunities. Journal of Financial Perspectives, 4(1). Communications on Applied Nonlinear Analysis ISSN: 1074-133X Vol 29 No. 4 (2022) 25 https://internationalpubls.com [3] Madakam, S., Holmukhe, R.M. and Jaiswal, D.K., 2019. The future digital work force: robotic process automation (RPA). JISTEM-Journal of Information Systems and Technology Management, 16, p.e201916001. [4] Andrade, D., 2020. Challenges of automated software testing with robotic process automation RPA-A comparative analysis of UiPath and automation anywhere. International Journal of Intelligent Computing Research (IJICR), 11(1), pp.1066-1072. [5] Jeyaraj, A. and Sethi, V., 2020. Embedding Robotic Process Automation into Process Management: Case Study of using taskt. AIS Transactions on Enterprise Systems, 5(1), p.1. [6] Jha, R.S. and Sahoo, P.R., 2020. Internet of things (IoT)–enabler for connecting world. In ICT for competitive strategies (pp. 1-7). CRC Press. [7] Axa.com, 2020, Overview of the Company, Available at: https://www.axa.com/ (Accessed on: 31 March 2021). [8] Rai, D., Siddiqui, S., Pawar, M. and Goyal, S., 2019. Robotic process automation: the virtual workforce. International Journal on Future Revolution in Computer Science & Communication Engineering, 5(2), pp.28-32. [9] Zurich.ch, 2020, Overview of the Company, Available at: https://www.zurich.ch/en/private- customers/provision-assets-life/risk-life- insurance?pat=71700000119998314&gad_source=1&gclid=CjwKCAjw5PK_BhBBEiwAL 7GTPbIS1UoSm_jtNi_jYuODYOeIHUn8E-kLkeDFJK27M34WlK_Mo5- 8ZhoCZlIQAvD_BwE&gclsrc=aw.ds (Accessed on: 14th April 2021). [10] Eckert, C. and Osterrieder, K., 2020. How digitalization affects insurance companies: overview and use cases of digital technologies. Zeitschrift für die gesamte Versicherungswissenschaft, 109(5), pp.333-360. [11] Allianz.com, 2020, Overview of the Company, Available at: https://www.allianz.com/en.html (Accessed on: 10th April 2021). [12] Qaudeer, M., 2020. Blockchain And Its Transformational Impact To Global Business. [13] Asenahabi, B.M., 2019. Basics of research design: A guide to selecting appropriate research design. International Journal of Contemporary Applied Researches, 6(5), pp.76-89. [14] Techhq.com, 2019, AXA saves $182K in six months with RPA, Available at: https://techhq.com/2019/02/axa-saves-182k-in-six-months-with- rpa/#:~:text=AXA%20saves%20$182K%20in%20six%20months%20with,result%20of%20 deploying%20robots%20in%20its%20UK (Accessed on: 7th April 2021). https://www.axa.com/ https://www.zurich.ch/en/private-customers/provision-assets-life/risk-life-insurance?pat=71700000119998314&gad_source=1&gclid=CjwKCAjw5PK_BhBBEiwAL7GTPbIS1UoSm_jtNi_jYuODYOeIHUn8E-kLkeDFJK27M34WlK_Mo5-8ZhoCZlIQAvD_BwE&gclsrc=aw.ds https://www.zurich.ch/en/private-customers/provision-assets-life/risk-life-insurance?pat=71700000119998314&gad_source=1&gclid=CjwKCAjw5PK_BhBBEiwAL7GTPbIS1UoSm_jtNi_jYuODYOeIHUn8E-kLkeDFJK27M34WlK_Mo5-8ZhoCZlIQAvD_BwE&gclsrc=aw.ds https://www.zurich.ch/en/private-customers/provision-assets-life/risk-life-insurance?pat=71700000119998314&gad_source=1&gclid=CjwKCAjw5PK_BhBBEiwAL7GTPbIS1UoSm_jtNi_jYuODYOeIHUn8E-kLkeDFJK27M34WlK_Mo5-8ZhoCZlIQAvD_BwE&gclsrc=aw.ds https://www.zurich.ch/en/private-customers/provision-assets-life/risk-life-insurance?pat=71700000119998314&gad_source=1&gclid=CjwKCAjw5PK_BhBBEiwAL7GTPbIS1UoSm_jtNi_jYuODYOeIHUn8E-kLkeDFJK27M34WlK_Mo5-8ZhoCZlIQAvD_BwE&gclsrc=aw.ds https://www.zurich.ch/en/private-customers/provision-assets-life/risk-life-insurance?pat=71700000119998314&gad_source=1&gclid=CjwKCAjw5PK_BhBBEiwAL7GTPbIS1UoSm_jtNi_jYuODYOeIHUn8E-kLkeDFJK27M34WlK_Mo5-8ZhoCZlIQAvD_BwE&gclsrc=aw.ds https://www.allianz.com/en.html https://techhq.com/2019/02/axa-saves-182k-in-six-months-with-rpa/#%3A~%3Atext%3DAXA%20saves%20%24182K%20in%20six%20months%20with%2Cresult%20of%20deploying%20robots%20in%20its%20UK https://techhq.com/2019/02/axa-saves-182k-in-six-months-with-rpa/#%3A~%3Atext%3DAXA%20saves%20%24182K%20in%20six%20months%20with%2Cresult%20of%20deploying%20robots%20in%20its%20UK https://techhq.com/2019/02/axa-saves-182k-in-six-months-with-rpa/#%3A~%3Atext%3DAXA%20saves%20%24182K%20in%20six%20months%20with%2Cresult%20of%20deploying%20robots%20in%20its%20UK Communications on Applied Nonlinear Analysis ISSN: 1074-133X Vol 29 No. 4 (2022) 26 https://internationalpubls.com [15] Thelabconsulting.com, 2018, RPA in Insurance - Robot Process Automation in Insurance Claims Processing, Available at: https://thelabconsulting.com/rpa-in-insurance-robot- process-automation-in-insurance-claims-processing/ (Accessed on: 9th April 2020). [16] Generali.com, 2020, Overview of the Company, Available at: https://www.generali.com/ (Accessed on: 19th April 2021). [17] Mahaputra, M.R., 2017. The influence of trust and customer value to customer satisfaction on bank BRI Branch Soetomo Jambi. Saudi Journal of Business and Management Studies, 2(8), pp.737-743. [18] Unit8.com, 2021, A Guide to Building a Financial Transaction Anomaly Detector, Available at: https://unit8.com/resources/a-guide-to-building-a-financial-transaction-anomaly-detector/ (Accessed on: 15th May 2021). [19] Andersenlab.com, 2021, RPA in Insurance: Why Automation Is Inevitable and How to Introduce It, Available at: https://andersenlab.com/blueprint/introduce-robotic-process- automation-insurance [20] Nanalyze.com, 2021, UiPath is the Latest Robotic Process Automation Stock, Available at: https://www.nanalyze.com/2021/03/uipath-robotic-process-automation-stock/ (Accessed on: 12th April 2021). [21] Le Clair, C., Cullen, A. and King, M., 2017. The forrester wave™: Robotic process automation, q1 2017. Forrester Research, 770. https://thelabconsulting.com/rpa-in-insurance-robot-process-automation-in-insurance-claims-processing/ https://thelabconsulting.com/rpa-in-insurance-robot-process-automation-in-insurance-claims-processing/ https://www.generali.com/ https://unit8.com/resources/a-guide-to-building-a-financial-transaction-anomaly-detector/ https://andersenlab.com/blueprint/introduce-robotic-process-automation-insurance https://andersenlab.com/blueprint/introduce-robotic-process-automation-insurance https://www.nanalyze.com/2021/03/uipath-robotic-process-automation-stock/ Praneetha Kotla I. INTRODUCTION Figure 1: Aim of Transformation II. LITERATURE REVIEW Figure 2: Stages of information value loop III. METHODOLOGY Figure 3: Evaluation matrices IV. RESULTS Figure 4: Distribution of Anomaly Scores in Financial Transactions Figure 5: Technological Innovation among Insurers Figure 6: UiPath Deployment Table 1: Case Study Analysis Table 2: Comparative Analysis V. DISCUSSION VI. CONCLUSION AND FUTURE WORK VII. REFERENCE LIST