We develop a model and a solution technique for the problem of generat
ing electric power when demands are not certain. We also provide techn
iques for improving the current methods used in solving the traditiona
l unit commitment problem. The solution strategy can be run in paralle
l due to the separable nature of the relaxation used. Numerical result
s indicate significant savings in the cost of operating power generati
ng systems when the stochastic model is used instead of the determinis
tic model.