M3 Master LEP

In the initial stages of development, one key challenge our team faced was designing a user interface (UI) that effectively accommodates the input of constraints and preferences from teachers. None of us is a professional UX-UI designer, therefore, creating an appealing and simple UI for admin or teachers to specify their requirements posed a significant hurdle, as it required striking a balance between complexity and user-friendliness. We undertook several brainstorming sessions ensuring that the UI could be used efficiently without overwhelming the user.

One of the first problem in our development journey revolved around efficiently reading and importing the data stored in Excel files into our database. Handling data migration from Excel to a database format presented a unique set of challenges. Throughout the project implementation, further key challenges included establishing effective storage solutions within our database, defining data structures and relations, and ensuring the data integrity and consistency. Addressing these hurdles was essential for a successful project outcome, and the devised solutions significantly enhanced the effectiveness of our data migration process.

In our student project, the main challenge lay in developing a time planning system capable of generating a schedule that meets all our constraints. Our search for an open-source solution led us to Timefold, a promising library for automated schedule generation. However, integrating Timefold into our project posed its own set of challenges, such as aligning our data structure and code implementation with Timefold’s functionality. The adaptation process required in-depth comprehension of Timefold itself. The library proved to be transformative, becoming the backbone of our project.

Experimenting with different solver configurations and approaches to solve the planning issue also consumed a considerable amount of time. We tried numerous configurations in an effort to optimize the system, but these attempts were time-consuming and often did not yield the expected improvements in schedule optimization. In overcoming these challenges, we not only technically integrated Timefold but also elevated the efficiency and effectiveness of our time planning system.

Implementing defined constraints was a notable challenge during the development phase. The specified constraints were initially documented in a human-readable manner, and then our focus shifted to transforming them into a format compatible with Timefold for scoring solutions. Timefold operates with constraints structured in a stream-like fashion, introducing an additional layer of complexity to the translation process. The successful implementation of the constraints allows us to generate a schedule that meets all the predefined constraints and to ensure the feasibility of the planning.

Security was also a priority in our project, especially because we dealt with sensitive personal information. To safeguard the stored data in MongoDB and prevent any potential data leaks, we implemented an approach involving the utilization of Java Cryptography Extension (JCE) and Java Cryptography Architecture (JCA) for data encryption and decryption. The task became even more complex when we added the Login feature, which required the implementation of JSON Web Token (JWT) for secure authentication and configuring Cross-Origin Resource Sharing (CORS) to allow only known sources and block unauthorized attempts. One of the most challenging parts was integrating everything smoothly with Spring Boot while adhering to its guidelines and conventions. Despite these challenges, our team successfully addressed these complexities, including the seamless integration of JWT for login authentication and ensuring that the data is stored in an anonymized format, establishing a secure foundation for the project’s functionality. This dedication to data security ensures a safe and confidential environment for the users of our platform.