Research Aims
Continuous integration is used in many Devops projects. Use it with Machine Learning, during the process, give the control of the code and the continuous release of the project’s target huge improvement, and also give decisions to Project Manager help them find out what’s the best conditionat the stage.Bring lots of benefits to tee final product submission of the project. Does continuous integration bring only software-level improvements in such projects? How can it to be used as a Project Manager tools to make decisions, go a step further can it make change for each project management auto as it need to be? For project management, what kind of decision made by machine can be brought about at each stage of project control? The main purpose of this research is to study machine learning and the impact on the management of such projects. How can human combine with machine make decision? Can it only with machine to make decisions in each stage of project process? What benefits can the research benefit from the project itself, how much improvement can be brought to the project management, and at which stage of the project management will have an impact and impact? What are the improvements in project control?
Background
Continuous Integration (CI) has become a disruptive innovation in software development: with proper tool support and adoption, positive effects have been demonstrated for pull request throughput and scaling up of project sizes. As any other innovation, adopting CI implies adapting existing practices in order to take full advantage of its potential, and “best practices” to that end have been proposed. My master thesis has done some research on the impact of CI on the Pipeline Engineering Information System project, and got a master thesis, which provided some help with the CI model I have made for the progress control of the information system project. Combining the existing projects, give some decisions in each project process to project manager using Devops to control project progress, the use of CI will be natural. During this period I am thinking why don’t we give method and key values to Machines and let them export results and change by themselves for the project management? What I am studying is with machine learning to study on a area with project management, use particular tools, CI, to influence the Project Management process at each stage. If not used, I can try to build a Devops model, experiment with the project management control process, and compare it with the progress of the existing project to help in the project control process. The current progress is compared. Why can I use CI? I can participate in the research in this project.
Case Study
Software-Defined Mobile Supply Chains, during the work on this project, establish research models and process control understanding, help team to make the goal with each milestone, with help for the team get more information with it, work insight into the process of project management, set up Devops model with programmer and Project Manager, get relevant data and write the corresponding paper within the specified time.
Research Questions
The research will seek to answer the following questions:
- How can CI with Machine Learning to be done in the software development process?
- To whom need decisions with this model?
- What are the most important process can use during whole Project life cycle?
- What activities or services engage in or provide, and what is the impact of CI in the Project Management?
- What are the benefits of, and what infection when use for the Project?
Methodology
The proposed study will be conducted bearing in mind issues of costs, time and access to Machine Learning, Software-Defined Mobile Supply Chains. It will be carried out in three phases, using quantitative and qualitative methods to collect primary data.
- Phase One will involve an in-depth literature review to achieve a more detailed theoretical framework and definitions of variables that will influence questionnaire and interview designs and subsequent analyses.
- During Phase Two, primary and secondary data will be collected from various identified data during the software development process. Primary data will be collected using CI model auto collected data, interviews and questionnaires. Ci model will separated from production environment, these interviews and questionnaires will be administered in my working team. Secondary data sources will include books, journals, reports, statutes and archival information.
- During Stage Three, the data generated and collected will be analysis and interpreted to come out with the research findings.
Research Challenges
Set up Machine Learning environment with Devops in a proper Software Development project, with proper team can be done, but most of Devops project can’t be executed in its expected. Some problems that could affect the research include a limited access to some part of development team like the security defend coding team. Make a separate model with Machine Learning combine with security defend test can solve it. Another problem that could arise will be the transmission of inherent errors and biases of data collected from secondary sources. To resolve this, secondary data that will be used will be limited to official documents and research done by appropriately recognized bodies who would have likely taken the necessary precautions to reduce such errors and biases.
Provisional Timeline
Phase One and questionnaire design will be completed by the end of June 2019 and Phase Two will be conducted in rest of 2019 and full 2020. Phase Three and a first draft will be completed by the end 2021. The final working paper will be submitted to TU Dresden by the end of January 2022.
By Gavin Wang in 14.05.2019.
It can be copied, but please add the Original Writer Name.