Yes, the emergency triggers are the only thing, that can be trusted during difficult time. But there are many issues related to them. i.e in case of autonomous responder, it needs to accurate and precise with intelligent decision making approach. You gave an example of ship accident, we have an example of very famous movie titanic. the context of that movie tells us the rescuers helped the rich people first and boarded them off the ship. This influence tells us that passengers cannot trust fully the decision making or path finding approach by a human. In case of computers, through testing is required to get accurate results. 

yes you can

Parvaneh, it depends on why you're asking.  Functionally, one person can serve another person in a wide variety of roles, including helping with decisions. There is a rich literature in the organizational psychology, cognitive psychology, human factors, and economics literatures that describe this. Do they meet some definition of "decision support?" Not typically. The academic literature that uses the term "Decision Support" is specifically focused on software algorithms and interfaces that help with decisions. While, in some cases, a human could perform the same role as the software that process of a human performing the role wouldn't be interesting to that academic community, except maybe as an example of expert behavior.  

Getting back to your question... why are you asking? Are you looking to use concepts from the decision support community to describe / taxonomize human behavior (e.g. the behavior of emergency responders)? That could make sense.  Or are you looking to show how computational decision support methods are part of a larger socio-technical system and can/should be supplemented by human activity? Also a valid approach.  Also, as I suggested above, doing analysis of specific human activity as a motivator for new software algorithms / interfaces is also valid. 

On the other hand, I'd be careful using the term "decision support" metaphorically as a short hand for "one person helping another person with a decision."  That's not what the term means to the decision support community and will be more confusing than helpful. Unless, of course, there is another community that uses the term "decision support differently."  I'm only familiar with the term from the computational decision support community. 

Inside of the theory of decision analysis, a young movement is the naturalistic decision making that I think that is more adequated to deal with your kind of decision. It is based on:

• an uncertain and dynamic context (the focus is on real-world situations, without the simplification of evaluations in the laboratory),
• for the solution of ill-structured problems (real problems without a complete definition, as a firefighter facing the fire in a closed environment),
• inside of unstable norms and organizational objectives (they are poorly defined or are competitive or change).
• action / chain reaction: in contrast with unique decisions;
• time constraints: there is no time to generate multiple alternatives and to compare them;
• High impact: situations where the decision has a major impact for the participants of the process;

Participants are  experts in the application area in which you are making the decision, know what needs to be done and what goals need to be achieved.

Zsambok, C. E., & Klein, G. (2014). Naturalistic decision making. 2nd edition. Psychology Press

Klein, G. (2008). Naturalistic decision making. Human Factors: The Journal of the Human Factors and Ergonomics Society, 50(3), 456-460.

Lipshitz, R., Klein, G., Orasanu, J., & Salas, E. (2001). Taking stock of naturalistic decision making. Journal of behavioral decision making, 14(5), 331-352.

Lipshitz, R., & Strauss, O. (1997). Coping with uncertainty: A naturalistic decision-making analysis. Organizational behavior and human decision processes, 69(2), 149-163.

Thank you so much everyone for your useful answers. I really appreciate it.