Context. In practice, software projects frequently fail in many fields, which causes the huge loss for the human being. Assumption faults are recognized as a main reason for the software project failures. As the world is changing fast, environment assumptions of software can be easily wrong. The daily assumption-related activities show not enough effectiveness and efficiency to deal with assumption faults. For example, no documenting of key assumptions, inappropriate assumption validation, lack of knowledge. In research, there is no empirical research about assumption management practice. Two assumption management frameworks were outlined. They both support the assumption formulation and assumption management. The formal assumption management framework provides an assumption-component mapping function to analyze assumption failures. Objectives. Our goal is figuring out how development team members handle environment assumptions today in practice and how they might handle them better tomorrow. To be specific, I test the applicability of the so far theoretical assumption management frameworks and investigate the assumption type, assumption formulation and assumption management in practical software development Methods. An interview-based survey was implemented with 6 product managers from Chinese software companies. They have rich experiences on assumption management and software development. I used directed content analysis to analyze the qualitative data. The result of the research is intended to be a static validation of the assumption management frameworks. Results. Interviewees consider that the assumption-component mapping function of the formal assumption management framework is useful in making decisions and analyzing the problems. However, using these frameworks takes too much effort. The functions of frameworks are covered by the development team members and the existing tools. Assumptions tend to be discovered when they frequently change and are important to the requirements. The main assumption types are user habit assumptions and quality attribute assumptions, which are both requirement assumptions. The user habit assumptions consist of name, description and value, while the quality attribute assumption formulation is name and value. The major assumption treatment activities are figuring out the value of assumptions, assumption monitoring, assumption validation and handling assumption failures. Assumption failures result in the loss of users and benefits. Assumption failures are always caused by the poor ability and experience of development team members. Conclusion. I create an assumption management model based on my result, and find out the advantages and disadvantages of the formal assumption management framework and semi-formal assumption management framework. The research could help improve the efficiency and effectiveness of assumption management practice. Also. The research can be treated as the starting point to study assumption management practice deeper.
Identifer | oai:union.ndltd.org:UPSALLA1/oai:DiVA.org:bth-15406 |
Date | January 2017 |
Creators | Zhang, Guangyu |
Publisher | Blekinge Tekniska Högskola, Institutionen för programvaruteknik |
Source Sets | DiVA Archive at Upsalla University |
Language | English |
Detected Language | English |
Type | Student thesis, info:eu-repo/semantics/bachelorThesis, text |
Format | application/pdf |
Rights | info:eu-repo/semantics/openAccess |
Page generated in 0.0018 seconds