University of Twente Student Theses


Deciding between mobile implementation platforms - A multi-criteria decision system for implementation platforms for mobile applications

Fluttert, Ernst (2013) Deciding between mobile implementation platforms - A multi-criteria decision system for implementation platforms for mobile applications.

[img] PDF
Abstract:The current mobile market is segmented into several mobile operating systems: iOS by Apple, Android by Google, WindowsPhone by Microsoft, and BlackBerry by RIM. Each mobile operating system has its own software development kit (SDK), programming language, interface guidelines, and review process. To create a mobile app that works on all of these platforms, called cross-platform apps, there are roughly three options: native apps, web apps, and hybrid apps. The first option would be to develop a native app for each OS that the organization wants to, which is basically developing one app, four times. The web app option is to develop a mobile-optimized website that can be accessed using a browser, independent of the platform. The last option, hybrid app, uses a wrapper that can extend a web app with native functionality and native distribution. This exploratory research has several goals. The first is to determine criteria for mobile apps, and their relevance towards each implementation platform (native, web, or hybrid). The second goal, and deliverable of this research, is to develop a decision support system to determine the most favorable mobile implementation platform based on a set of criteria. The final goal is to validate this tool on correctness and practical appliance. In the conclusion this answers the main research question: “How to transform mobile application criteria into decisions for mobile implementation platforms using a decision support system?”. The mobile app criteria were determined by interviewing experts, which was filtered by applying methods from qualitative data processing. Each interview consisted of questions about business drivers, technical requirements, and any other aspect that related to developing mobile apps. The result is a list containing 16 criteria, where the most relevant are: 1) supporting multiple mobile OS’s; 2) offline usage; 3) utilizing device sensors; 4) distribution via application store; 5) require native functionality; 6) type of content. The foundation for the decision support tool is multiple-criteria decision making (MDCM). This technique evaluates multiple criteria to select the best alternative based on given input set. Two MDCM types were investigated: multi-criteria value function and analytical hierarchy process. The multi-criteria value function is a simple method where each criterion has a weight and the highest scoring alternative is the best. Analytical hierarchy process also includes priority of the criteria, which consumes more time but also gives a finer granularity than the multi-criteria value function. The final prototype was based on a multi-criteria value function, as the tool should be used as an indicator. The prototype was tested by a case at a client and further reviews by experts. This test confirmed that the tool operates well in practice; but also needed some refinement in criteria definitions and weights. Further insight in the outcomes of the tool was done by a sensitivity analysis which calculates all possible combinations of criteria. This revealed that the current set of criteria strongly favor native implementations of apps. Other results of this study are that the prototype is a very practical tool, but has to be updated frequently to keep up with the latest technology developments in the mobile ecosystem. The linearity of the model helps clients to understand how a criterion influences the alternatives. The tool gives a score to each platform and it could happen that these are in close range of each other, which also gives room for interpretation of the result.
Item Type:Essay (Master)
Faculty:EEMCS: Electrical Engineering, Mathematics and Computer Science
Subject:54 computer science
Programme:Computer Science MSc (60300)
Link to this item:
Export this item as:BibTeX
HTML Citation
Reference Manager


Repository Staff Only: item control page