| dc.description.abstract | The development of a real-time “E-Shuttle” campus shuttle tracking system poses a
software engineering challenge due to the need to integrate complex GPS based
algorithms, continuous real-time data processing, and highly usable mobile and
web interfaces within a single platform. Such systems require both rigorous process
correctness in early engineering phases and flexibility to support rapid user interface
evolution, yet there is limited guidance on selecting a development methodology
that effectively balances these competing requirements. The objective of this study
is to identify the most suitable software development methodology for the E-Shuttle
system and to determine whether a single methodology or a hybrid approach is more
appropriate. Seven major development tasks were defined based on the functional
and non-functional requirements of the system and were comparatively evaluated
using Agile development and the Predictive-cum-Adaptive Process-centric Development
to User-centric Tool methodology. A task-level weighted evaluation was conducted
using four criteria: ease of execution, time efficiency, project suitability, and ability to
handle constraints and technical complexity. The results indicate that the Predictive cum-Adaptive Process-centric Development to User-centric Tool methodology is more
effective during early engineering activities such as requirements elicitation and system
architecture design, where detailed process modelling and constraint analysis are
essential. In contrast, Agile development demonstrates superior performance during
implementation, integration, testing, deployment, and maintenance phases, where
iterative development and rapid stakeholder feedback are critical. The study concludes
that no single methodology adequately supports all phases of the E-Shuttle project
and that a hybrid approach provides the most effective solution. By applying a
process-centric methodology in early analytical phases and an Agile approach in later
development and operational phases, the proposed hybrid methodology offers both
architectural stability and development flexibility, contributing practical guidance for
methodology selection in real-time campus transportation systems and similar complex
software projects. | en_US |
