Attendance systems based on mobile applications have been widely implemented, but they often lack precise location determination, including the absence of altitude parameters to differentiate between floors in multi-level buildings. This limitation can lead to inaccurate attendance records and potential fraud. Therefore, a more precise attendance system is needed, one that can accurately determine both horizontal and vertical (altitude) positioning to support transparent and accurate attendance tracking.

This system was developed using the prototyping method, which allows continuous interaction between developers and stakeholders through iterative prototypes. Each version was tested for core features such as GPS- and barcode-based attendance, location management, leave approval, and attendance requests. This approach is effective for complex systems requiring precise location validation, including accurate altitude readings. The technology stack includes Next.js for the web frontend, Flutter for the mobile app, and PostgreSQL as the relational database, consisting of 14 main tables covering users, companies, attendance, leave, requests, and locations.

The test results show that the application performed 100% as expected based on black box testing scenarios. A total of 14 test schemes were designed to evaluate all the main system features. The system is capable of distinguishing attendance locations with high accuracy, including a maximum tolerance radius of ≤8 meters and consistent altitude readings. All 14 main features function properly according to the needs of each user role, indicating that the system successfully meets all functional aspects for precise and real-time attendance recording and monitoring.