Our organization is currently utilizing Azure DevOps (ADO), and we're interested in integrating the "Team Data Science Process" (TDSP) into our data science and machine learning project workflows within ADO.
Upon accessing ADO Board --> Process, I notice that there are four available processes:
What I am seeking guidance on is how we can effectively incorporate TDSP using one of these processes. Specifically, I would like each new project to encompass the following stages, each of which can be labeled as either 'To Do', 'Doing', or 'Done':
Any assistance or advice on how we can achieve this would be greatly appreciated.
I would like each new project to encompass the following stages, each of which can be labeled as either 'To Do', 'Doing', or 'Done'
All processes support the above states, 'To Do', 'Doing', or 'Done'. It is recommended that you read Default processes to have a basic understanding of each process and choose the one that suits you.
Among the available processes, Agile or Scrum would be more suitable for integrating TDSP due to their flexibility and iterative nature.
You can use different work item types to represent your different stages.
You can create inheritance process based on Agile or Scrum .
Add custom work item types to reflect the stages of TDSP. For example, for Business Understanding, create a work item type called "Business Understanding", and so on.
For each new work item type, define the states as 'To Do', 'Doing', and 'Done'. This will help you track the progress of each stage.
You can also use custom states to represent the stages without create new work item type.
The above suggestions are just to provide you with some ideas, they may not fully meet your situation. Adjust them according to your actual needs.