8

This has already been asked (How can a team working on research based items work in an Agile project?), but the thread is from 2011, so I thought I'd poke for an update.

Does anyone know of a good process for managing research projects or a whole research organization (as in scientific research rather than R&D) through Agile methodology? Does anyone know examples of institutes doing this?

I was able to find SCORE (Scrum for Research) but it's mainly for teaching faculty and research students. In a more recent piece Agile Research the authors conclude that "no “academic agile process” has yet been laid out". They do take the subject a bit further, developing "the principles of agile research" based on their software development counterparts. I've made a preliminary search for scientific papers on agile research, but with no results worth mentioning.

All in all, there seems to be a surprising lack of documents describing the use of Agile in research. Why? It's not a novel idea - has it ultimately proven to be a bad idea?

  • Given that scrum was in fact inspired by techniques employed in Polymer reasearch at DuPont, someone will probably have something for you. – Nathan Cooper Sep 22 '15 at 11:21
  • If you consider "Iterative and incremental" as agile (without any kind of trademark on the name), it's been around for a long time and has been used in scientific research projects. See this great article about the history of iterative and incremental development citeseerx.ist.psu.edu/viewdoc/… – Fuhrmanator Nov 8 '17 at 16:32
  • I'd also add that there's a paradox in that traditional funding organizations for scientific research tend to be pretty waterfall in nature. The academic career model (university tenure) is largely based on getting publications, which allow a researcher to get more funding via proposals (promises to research funders), to allow hiring students and buying equipment to do research that will result in publishing more results, etc. That "machine" is pretty far from agile... Nothing stops a research organization from executing a research project in an agile fashion, however. – Fuhrmanator Nov 8 '17 at 16:45
  • Also, a very interesting read, which details the "publish or perish" ecosystem: scrum-breakfast.com/2014/04/agile-scientific-research.html – Fuhrmanator Nov 8 '17 at 18:43
5

I think the reason you've not found much written is that when you get down to the cores of agile, the difference in projects is not all that pronounced. What you want from a project is what really drives the type of implementation, not what you are creating.

Predictable Schedule, Flexible Scope- Use Scrum

Volatile scope, short delivery windows- Use Kanban

Fixed Scope, Flexible Schedule- Use Kanban or Scrum

End product completely unknown- Use XP or Lean Startup

Based on past advice I've given on school projects and some research stuff, I suggest trying a Kanban workflow. You have a good idea what you need to do at the start. However, as you move through the research project, findings will likely change your direction or scope. Working in a Kanban flow allows easy change of the backlog.

If you truly don't know what your research results will bring and you are trying to develop something marketable, then the Lean Startup Build, Measure, Learn cycle may be helpful.

Hopefully that was enough detail without turning this into an in depth dissertation.

2

I was faced with the same question recently, and have to say, finding anything remotely directive in approach is very very hard. The closest I found as far as some guidelines on what to look at and how to begin was in this short article:

Agile Kanban for Research Work

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.