Although the existing road management system had operated reliably for many years, it was built as a monolithic application on a legacy ASP.NET Web Forms foundation, with business logic tightly coupled to the user interface. As a result, identifying the scope of impact for feature updates or defect investigations was difficult, creating a major obstacle to rapid feature additions and performance improvements in response to regulatory revisions and operational changes.

In addition, 54 shared master data sets and APIs were referenced across multiple subsystems, creating a constant risk that even a small change could affect unexpected functions. To ensure stable operations and support future DX initiatives, it had become essential to gradually migrate to a modern, highly maintainable backend platform without interrupting existing road management operations.

The main objective of this project is to migrate the core backend and APIs to the latest .NET 8 platform, which is covered by long-term support (LTS), while maintaining day-to-day road management operations, thereby ensuring scalability and maintainability for the future.

Our goal was not simply to replace the programming language or framework, but to achieve safe and reliable modernization that fully preserved the complex business logic accumulated over many years, master data integrity, and integration specifications with the front end.

SMILE handled the entire process end to end, from analyzing the existing system code and implementing the migration to the new architecture through quality verification and production release support. The main scope of work was as follows.

• Structural analysis and reverse engineering of an existing ASP.NET Web Forms system
• Migration and redesign of the backend API and core business logic to ASP.NET Core (.NET 8)
• Optimization of the data access layer and processing logic to fit the existing database structure
• Data migration, CRUD processing, and consistency validation for 54 shared master data sets
• Validation of API interface consistency with the React 18-based web frontend
• Architecture restructuring with clear separation of responsibilities from the ArcGIS (geographic information system) integration module
• Comprehensive regression testing, performance tuning, and production release support

Because this was a large-scale business system, we avoided the risks of a full Big Bang replacement and proceeded with a phased migration while visualizing the impact scope by function and API. We carefully analyzed undocumented implicit specifications embedded in the legacy code and accurately migrated them to the new platform.

In particular, we conducted a detailed review of the reference relationships between each screen and external process for the 54 shared master data sets that form the foundation of the entire system. Beyond simply passing data between components, we standardized input validation and error handling, creating a structure that maintains compatibility with the existing system while significantly improving future development efficiency.

SMILE also carefully verified the consistency of API response formats and communication protocols when integrating with the responsive web frontend built with React 18. Through repeated regression testing, we completed the backend modernization without compromising the user experience or responsiveness of the UI.

The objective was to extract the specifications and rebuild the system into a modern, loosely coupled architecture. In particular, 54 shared master data sets and ArcGIS-related geospatial data processing were deeply intertwined, creating a risk that even minor changes could cause widespread regressions.

To address this, SMILE created a functional dependency matrix early in the migration to quantitatively visualize the degree of coupling between APIs and shared master data. Based on this analysis, we clearly separated responsibilities between ArcGIS-dependent map processing and the general road ledger and business management logic, creating a more loosely coupled architecture. In addition, by rigorously applying a validation cycle of unit testing, API integration testing, and regression testing for each migration target, we eliminated potential defects before they could affect existing operations.

By reviewing data access processes and optimizing asynchronous processing with async/await, SMILE eliminated the causes of API response delays and improved processing performance across the entire system.

Migrating to the latest ASP.NET Core (.NET 8) platform eliminated the black-box nature of the legacy Web Forms foundation and dramatically improved the maintainability and scalability of the road management system.

Through comprehensive regression testing and phased migration, we completed the transition with zero impact on existing operations and no degradation. This eliminated long-standing concerns around system obsolescence and end-of-support risks, establishing a sustainable development platform that enables future feature enhancements and responses to legal revisions to be delivered quickly and cost-effectively.

The requirements and specification team in Japan worked closely with the development team in Vietnam, executing the project through a carefully designed communication framework. To address the challenges typical of legacy migrations, where past design documentation is often insufficient, the Bridge SE (BrSE) connected business requirements with the results of code analysis. Items requiring confirmation and potential risk factors were quickly visualized in Q&A sheets and impact scope lists.

To prevent discrepancies in understanding the specifications, we thoroughly prepared technical documentation in both Japanese and Vietnamese and shared test cases at an early stage to align expectations. Although the project was delivered through offshore development, we maintained the optimal balance of quality, speed, and cost, achieving a high-quality delivery on schedule.

For long-running core business systems, migrating from a legacy platform to a modern framework without interrupting operations is essential to successful corporate DX (digital transformation). In this case, SMILE provided end-to-end support from existing system analysis, API and business logic migration, and DB access adjustments to integrated verification of 54 shared master data sets and regression testing, achieving both stable operations and improved maintainability.

SMILE helps customers struggling with black-box legacy systems and technical debt achieve optimal modernization through proven technical expertise and meticulous reverse engineering.