Coffee Time is considering selling juices along with its other products.
                         States of Nature
                High Sales         Medium Sales         Low Sales
                   A(0.2)             B(0.5)               C(0.3)
A1 (sell juices)    3000              2000               -6000
A2 (Don't sell juices) 0               0                  0
The probabilities shown above represent the states of nature and the
decision markers's (e.g., manager) degree of uncertainties and
personal judgment on the occurrence of each state. What is the
expected payoff for actions A1 and A2 above? What would be your
recommendation? Interpret the results based on practical
considerations.

I searched +table +"expected value" +"states of nature"
and found this pfd file: http://www.cba.uni.edu/slides/920020_Mardis/risk.pdf

I know my english isn't good (it is not my mother tongue) but I hope
this could help you.

----------------------------------------------------------------------------

Expected payoff for actions A1 and A2

For each action, multiply the probability of each state of nature by
the sales value for that state.
The sum of that products is the expected payoff for each action.

E(A1) = 3000*P(A) + 2000*P(B) + (-6000)*P(C)
E(A1) = 3000*(0.2) + 2000*(0.5) + (-6000)*(0.3) = -200
Expected payoff for A1 is -200

E(A2) = 0*P(A) + 0*P(B) + 0*P(C)
E(A2) = 0*(0.2) + 0*(0.5) + 0*(0.3) = 0
Expected payoff for A2 is 0

----------------------------------------------------------------------------

I would recommend don't sell juices, because you should choose the
action yielding the best expected returns. In this case 0 is the
highest average return.

There are other decision rules and different approaches that could
take to different results (such as the optimistic approach, the
conservative approach. See:
http://homepages.stmartin.edu/fac_staff/dstout/BA302/QMB8ch04.ppt)