Decide what is more important in your study, detail or reliability. (See Reference 2.) If it is more important to include a rich description of a phenomena, such as thoroughly documenting how crisis responders worked during the 2011 Japanese earthquake and tsunami, then you may not be able to generalize the results of the discussion because you isolated a particular instance (crisis responding) and discussed it as a case study. If you approach crisis responding across all crises or all earthquake crises, your results may be generalizable.
Use a research method that has been deemed reliable in many similar studies. You will still be unable to say definitively whether your study is generalizable, but if you use a similar questionnaire or interview process in your study as other researchers used studying similar phenomena, you can hypothesize that the results may be generalizable. For example, if you use a similar questionnaire when you interview crisis responders in Japan as another researcher used in another crisis situation, you may make comparisons and discuss which aspects of the results are similar in other crisis situations.
Replicate your study in another situation if you are using a novel research design or a set of research tools that you developed yourself. Over time, you can compare results as you or other researchers use similar methods, and your method may become generalizable if other situations produce similar results.