Date

2021-9-10

Author

Hacaloğlu, Tuna

Metadata

Show full item record

This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

Item Usage Stats

527
views

240
downloads

Cite This

Software Size Measurement is a critical task in Software Development Life Cycle (SDLC). It is the primary input for effort estimation models and an important measure for project control and process improvement. There exist various size measurement methods whose successes have already been proven for traditional software architectures and application domains. Functional size measurement (FSM) being one of them attracts specific attention due to its applicability at the early phases of SDLC. Although FSM methods were successful on the data-base centric, transaction-oriented stand-alone applications, their applicability on the new generation software architectures are not studied well. Today software is frequently service based, highly distributed, message driven, scalable and having unprecedented levels of availability. In these architectures, ‘event’ concept largely replaces the ‘data’ concept. In this thesis, considering the significance of the event concept in today’s software systems, we explored the potential of an event-based software size measurement method. For this aim, we collaborated with 5 software organization and conducted multiple case studies. As a result of this research, it is seen that the proposed model produce promising results; “Event points” correlates well with effort. According to the findings of our study, it can be concluded that event as base counting unit can be used for measuring software size for both traditional and novel architectures, it is possible to perform a measurement without considering the data as a counting base, event-based effort estimation models yield acceptable error rates and prediction performance in the effort estimation models.

Subject Keywords

Yazılım büyüklük ölçümü, olay, COSMIC, durum çalışması, efor kestirimi

URI

https://hdl.handle.net/11511/93110

Collections

Graduate School of Informatics, Thesis