Empowering Innovation in Banking Insurance: Expert System for Information Management

Introduction

Over the years, financial institutions have faced and solved challenges to optimise information management and improve overall performance.¹ In such organisations, precise information management is required, accompanied by communication capable of dealing with new challenges, tasks and strategies.² In this situation, the use of new technologies such as expert systems becomes essential; they are designed to simplify processes that a human expert would find complex. An expert system is a stand-alone program that simulates the actions of an expert in a particular field. They are developed because human experts find it difficult to explain the rules they use to make judgments in the field.³

They also simulate the reasoning process that experts use to solve certain problems. On the other hand, they can be used by non-experts to improve their problem-solving skills.⁴

Thus, expert systems aim at optimising information management, which is a set of processes and technologies that can control, organise, support and access the information lifecycle, i.e. facilitate the storage, acquisition, recording and dissemination of information. This means that information becomes a resource of great strategic importance that can be used to achieve goals, improve decision making and generate knowledge, particularly in different areas.¹

Most research has shown that the development of expert systems (ES) for information management (IM) has the potential to have a positive impact on efficiency, accuracy and precision. Expert systems are therefore very important to organisations as an effective means of managing management plans.^5–7 According to the study carried out by,⁵ the effectiveness of Expert Systems was determined in the Information Technology Management System of the company Sion Global in Peru. They concluded that ES have the potential to be used by companies because they can control the quality of services throughout the process.⁵

On the other hand,⁶ want to understand the principles of expert systems and in turn provide the right communication and information for system development. As a result, they found that the tests performed coincided in correcting all the errors, so they were presented well and without any problems. In conclusion, they provide an opportunity to streamline processes and obtain accurate, timely and reliable information for future work, using technological tools to optimise time and resources.⁶

Over time, expert systems have been proven to provide effective and proven support and automation for a wide range of business decision challenges.^8–11 However, most ES are very complex and require a high level of technical expertise to develop and maintain, which can be a problem for companies that do not have the capacity to develop and maintain such systems.¹²

However, there is a need for further and more comprehensive research that analyses and verifies the productivity of ES in information management processes. This research paper aims to fill this gap through an analysis of the incorporation of ES for Information Management (IM), focusing on the implementation and contribution to the improvement of IM within the insurance company.

Further research is essential to analyse and confirm the impact of this tool in the area of Information Management (IM) and real-time decision making. The demand for these systems is driven by the increasing complexity of financial information, the need to simplify access to data and the urgent need for a centralised and up-to-date database for comprehensive and effective support.

This study contributes to provide practical, updated and relevant information on the implementation of ES for Information Management; streamlining processes, reducing waiting times, improving the user experience in real time and optimising the handling of large amounts of data within the insurance company.

In this sense, the objective of this research is to implement an Expert System to Improve Information Management in order to improve the quality of user service, optimising time and resources Lima, 2023.

Methods

In this section we provide a detailed description of the methods used in the development and operation of our software, which is designed to improve business decision making.

Operation

It was worked with an agile methodology, following the 5 phases into which this methodology is divided,^13,14 as shown in Figure 1.

Figure 1. Phases of the Scrum methodology.

Initiation phase

In this phase, a thorough analysis is carried out to identify the specific requirements of the expert system, including decision rules, knowledge logic and the specific needs of the end user. In close collaboration with domain experts and end users, the elements of the product backlog are prioritised according to their relevance and contribution to the value of the expert system. In addition, an overview of the product backlog was provided and the responsibilities of each member of the development team were clearly defined. This was done to streamline decision making. In addition, the charter was drawn up, covering the project objectives and expected results (Figure 2).

Figure 2. Assigning roles.

Implementation phase

During the sprint, the team focuses on the effective implementation of new rules, knowledge integration and continuous improvement of the expert system logic. Active communication and continuous collaboration between developers, domain experts and end users is essential to ensure that the system evolves according to expectations. The structure of the expert system was also defined, including the server architecture with a master database and mirror replication. In addition, minutes were taken at the start of each phase, to-do lists were drawn up and sprints were planned. Figure 3 also shows the architectural design of the expert system, with a connection and permissions via Azure Active Directory and a connection to the company database, where it provides a holistic view of the system’s operation in relation to all the components involved. The tasks of database design, dashboard creation, prototype development and implementation were carried out, culminating in the final dashboard.

Figure 3. Expert system architecture for information management.

Review and retrospective phase

A detailed review of the decisions made by the expert system during the sprint is carried out to assess their accuracy and effectiveness. User feedback is essential to adjust and improve the rules and logic of the system, ensuring effective alignment with end-user expectations. In addition, a thorough verification of the system was carried out by performing unit tests to identify possible errors in the code, as shown in Figure 4, which shows the code of the frame loading procedure that will be uploaded to the Integration Services catalogue. Acceptance testing was also carried out in conjunction with the customer to ensure that the application was approved. However, Figure 5 shows the database diagram, which visualises how the data and the relationships between them are structured in the storage system.

Figure 4. Frame loading procedure.

Figure 5. Expert system database for information management.

Launch phase

The team reflects on the results of the sprint, identifying what worked well and areas for improvement. Corrective actions are proposed to address any challenges identified, with the aim of optimising the process and continually improving the system’s ability to learn and adapt. In addition, as the sprints progressed, testing was conducted in the presence of end users, who were empowered to request adjustments to both functionality and interface if deemed appropriate. Figure 6 provides a visual representation of the software development process. At the end of the process, we were able to establish agreements and successfully complete testing, which allowed us to improve the efficiency of delivery times and the final minutes of each phase.

Figure 6. Flowchart of application development.

Unique features

This software has unique attributes that distinguish it from other existing solutions:

By detailing these unique methods and features, we provide a clear blueprint for the development and implementation of our software tool in the banking sector, increasing its replicability and usefulness.

Use cases

In this section we present the results obtained in the qualifications in terms of achievement of learning objectives, level of understanding and ease of use; we also specify that the system data will be downloaded automatically once it has been located using the link found in the repository.¹⁵

Use case 1: Presentation of the dashboard

Figure 7, shows the “Administrator” user interface, which displays the Expert System dashboard, which provides a summarised and visually appealing view of key information related to the information management of a set of products, risks and formats. The features displayed on the dashboard are the following: a) Product chart, for a given period. b) Risk list chart, for a given period. c) Format chart, for a given period. d) Last period loaded per product, information per period.

Figure 7. Main menu administrator user interface.

Input: Access to the Dashboard view.

Output: Report of loaded graphs and periods.

Use case 2: Product registration, risk and frame loading process

Figure 8, illustrates the process of registering products (a), risks (b) and loading frames (c) for a new product. The first step is to search for the product by name, which is automatically completed by industry, type of insurance, channel, risk, type of structure and format. You then select whether the product should remain active or not. Then, in order to load the frame, it is necessary to select the product, the year and the month in order to store it in the list of frames, where data such as the date of operation, the full name, the type of document, the document, the payment frequency, the start of validity, the end of validity, the product, the premium, the insured value, the credit number, the type of credit and the active status are recorded. Finally, the user can validate the information to proceed with the data export process.

Figure 8. Product registration, risk and frame loading.

Input: Access the products section, access the risk section, access the plot section.

Output: Report and list of frames.

Use case 3: Electronic Payment Voucher Report (CPE)

Figure 9, shows three filters, Product to select, Year and Month, to export the information from the CPE Voucher list of each record, as well as the data shown in Figure 9, such as the date of operation, type of identity document, identity document number, names, paternal surname, maternal surname, place of birth, premium, policy number, start of validity, end of coverage, product name and month. Finally, the user will be able to validate the information in order to continue with the data export procedure, in this way the user will receive all the visual information personalised and adapted to the voucher.

Figura 9. Electronic Payment Voucher Report – CPE.

Input: Access the SUNAT file export section.

Output: Report and list of vouchers.

Use case 4: Supervised Firm Voucher Reports (CES)

In Figure 10, the user with the Administrator role has the ability to generate a variety of reports related to the export of data according to the type of format, either Supervised Company Voucher (CES) or Electronic Payment Voucher (CPE), using different selection criteria such as product, year and month. These reports can be exported to the platform in different formats, such as excel, csv, txt and xlxs, and are processed by deleting duplicates and homologation, after which they are sent to the user to be managed by the Superintendency of Banking and Insurance (SBS), Sunat and Banco Pichincha. In addition, the administrator can carry out the corresponding configurations and operations so that the user can validate the information.

Figure 10. Supervised Firm Voucher Reports (CES).

Input: Access the SUNAT file export section.

Output: Report and list of vouchers.

Discussion

Figure 7, shows how the use of dashboards has a positive impact on the efficiency of the interface, facilitating visualisation by providing detailed information to improve decision making and optimise time. This resource provides a global view of key data, facilitating a deep and rapid understanding of the information, allowing analysis to have a significant impact. According to Refs. 16, 17, monitoring contributes significantly to achieving results, facilitating strategic decisions, identifying risks and promoting productivity improvements. These aspects are essential to ensure the competitiveness of a company. Moreover, this tool is known to be relevant in the financial sector, highlighting its importance in process improvement and informed decision making for information management. This finding is in line with previous research conducted by Ref. 18, which emphasises that dashboards are tools that allow the sharing and visualisation of important data of an entity, simplifying the development of decisions in data management. Not only do they facilitate the understanding of important information, but they also enable the sharing of relevant data and streamline critical decision making in financial and other sectors. Dashboards are therefore a valuable tool for improving information management by presenting data in a visual way. This feature allows them to make the most of data and play an essential role in decision making.^19–21 In summary, these findings highlight the importance of dashboards as critical tools that positively impact efficiency, decision making and quality of user service, making them essential resources for technology and finance professionals involved in business intelligence and data analytics.

The results shown in Figure 8, demonstrate the system’s ability to capture products, risks, data loads, etc., and to take corrective action to maximise results before the end of the month. Previously, this information was processed manually, which took longer to generate data and had a higher probability of error. This system provides technology and finance professionals with fast, centralised access to information, simplifying the management of large volumes of data within the insurance company. This feature is critical to ensuring proper user support and effective decision making; in general, according to Refs. 22, 23 they also mention process improvement in their research by demonstrating that the lack of use of information technologies limits the generation and dissemination of information, directly affecting its quality. These findings highlight the importance and benefits of having an expert system for information management, with the aim of improving both user service and effective communication between technology and financial professionals. According to Ref. 24, it is highlighted how information technologies are associated with productive processes to increase labour productivity, becoming a transformative element. Similarly,²⁵ describes how educational management processes are improved through the implementation of an expert system that contributes to improving management by automating processes and activities related to institutional planning and evaluation, streamlining the analysis of crucial information for decision-making.

Figure 9, illustrates the importance of data reports in the management of financial information, which are useful for monitoring results and achieving objectives. These reports are generated from filters applied by the administrator in relation to user support, and provide comprehensive details of the support provided on a daily basis. These findings are consistent with the research of,^26,27 which indicates that the lack of summary data results in increased query times and complicates analysis. By providing a complete view of actions taken, this tool improves communication and avoids duplication of effort. The implementation of expert systems in finance can have a positive impact on efficiency, team coordination and ultimately the quality of care provided. In addition,^28,29 state that data-driven reporting in information systems is now essential in the financial sector as it brings significant benefits to decision making, such as improved access to and tracking of information, integration of records from different episodes of care, ongoing coordination and support in justifying financial decisions. These results highlight the importance of reporting in data analysis, and how it contributes significantly to improving user care and coordination between technical and financial professionals.

The results shown in Figure 10, demonstrate the support provided by an evolving report; this perspective, combined with the seasonal behaviour of the financial system, enables management to make effective decisions to monitor results and achieve objectives. In addition, exporting reports on an ad hoc basis allows access to this information from any device, allowing staff to focus on core business tasks in the field, avoiding time spent on operational tasks in the office; this is evidenced in studies by Refs. 30, 31, which highlight the lack of management information systems to support decision making in companies in the financial sector due to the absence of management indicators. Similarly,³² in his research, addresses the lack of importance given by companies to data analysis and how this practice has a negative impact on the development of their activities. Finally,³³ analyses how banking institutions can improve their competitiveness by improving their internal processes through the incorporation of information technologies in decision-making.

Conclusions

This study successfully developed and implemented an expert system for information management in an insurance company, specifically in the area of life insurance underwriting. Previously, this company relied on manual records. Throughout this research, the challenge of modernising the process of data collection, security, peace of mind and customer protection was met and overcome in order to provide a more efficient customer service with a higher level of service.

Firstly, a dashboard in an information management system is a valuable tool that can improve data management and the quality of customer service by providing relevant information in an accessible and efficient manner. We have also seen a significant reduction in user errors and increased security in information management. In addition, an expert reporting system has significant benefits for financial staff, saving them time by providing detailed and structured information in a short period of time, making their job easier.

Ultimately, it can be seen that after the implementation of the expert system, information management has improved significantly, service times have been optimised and user satisfaction has increased, resulting in a significant improvement in the handling of information management.