Research And Implementation Of A Problem-Oriented Approach To Requirements Change

With the development of science and technology,computers have been widely used in every corners of our society.Software projects have moved from high-tech to public applications.Almost all kinds of modern work can not be separated from software products.Software products are needed from large-scale enterprises to everyone's daily life.Modern people are not only concerned about what the software can do,but also concerned about the quality of the software and its stability.The key to ensuring software quality is obtaining requirements accurately and responding to requirements changes.Requirements changes can affect most stages of a software project,such as design,coding,testing,etc,The later requirements change happens,the higher the cost.In a software development project,requirements are mentioned for the first time.A requirements change is crucial because it is related to the quality and cost of the software.At the same time requirements change affects software projects at all times,and internal and external factors lead to these changes to happen.In the process of software development,if we can not deal with these changes on time,they will lead to rising costs and declining software quality.Finally,these changes may lead to project failures.Therefore,it is of theoretical and practical significance to study problem-oriented approach to responding to requirement changes.The Problem Frames approach was originally proposed by Michael A.Jackson,a British scholar in software engineering.Its core theory can be expressed in K,S |-R,in which the K refers to the relevant context of the software,including hardware,people,etc,the S refers to the specification of the software,the R refers to user's requirements,and the symbol“|-“refers to satisfaction of the requirements.This method describes the problem in problem-oriented perspective by constructing problem diagrams and context digrams.It emphasizes the importance of adhering closely to the real world,which facilitates the project stakeholders' understanding and participating in the discussion of requirements.The purpose of this dissertation is to apply the theory of Problem Frames in order to response to requirement changes and to develop a requirement change management tool which can be applied to an actual project,thus combining theory with practice.The approach of this dissertation is based on the theory of Problem Frames K,S|-R in which,when any of the context K,the specification S,the requirement R changes,how to keep ensuring that satisfaction relationship(|-)is established.This dissertation introduces the concept of change points and task lists.A change point includes the overall changes of K,S,R,and a task list is a kind of domain properties attached to the domain in a problem diagram.By refining and decomposing requirements,details of requirements changes caused by the change points are recorded.The complexity of requirement changes is quantitatively calculated based on the number of the shared problem domains between requirements.In this dissertation,the formula of complexity and its imple4menting algorithm are given.According to the above theory,this dissertation introduces the development process of a requirements change management tool which can be applied to practical projects.On the basis of the Problem Frames,this tool enables imputting of requirements,transformation of inputted requirements into problem diagrams,and providing feedback to the tool users.It also provides a method of calculating the complexity of the requirements change based on the number of shared problem domains.Multiple changes may occur in the actual project for the same requirement,which may also lead to conflicts in the requirement itself.The tool can record the contents of each change and generate a state transformation tracing chain for a requirement.The key points of each requirement change are recorded by the chain and associated task list,in order to help find the location of the change that leads to the conflict.Finally,two cases are used to demonstrate the practicability of this tool.The first case is that the requirement itself is complex but the correlation between requirements is low.The second case is just the opposite:the complexity of each requirement is not so high but the correlation between requirements is high.The tool is applied to the above cases,and the results are compared with the results obtained after the actual project,which produces consistent results.This shows the usability of this tool.This dissertation starts with the research of the Problem Frames theory,combines the theory with actual requirements changes,analyzes changes by creating a task list,creates the state transitions of the causal chain by tracking the requirements,calculate the complexity of requirements change by analyzing the shared problem domains in problem diagrams.It makes some contribution to the combination of Problem Frames theory and its practical applications.