Real-Time Optimization Dynamic Programming Classroom lecture notes pdf
MASSACHUSETTS INSTITUTE OF TECHNOLOGY
Department of Civil and Environmental Engineering
Real-time optimization problems rely on decision rules that specify how decisions should
maximize future benefit, given the current state of a system. State dependence provides a
convenient way to deal with uncertainty. Some examples:
• Reservoir releases – Decision rule specifies how current release should depend on current
storage. Primary uncertainty is future reservoir inflow.
• Water treatment – Decision rule specifies how current operating conditions (e.g. temperature
or chemical inputs) should depend on current concentration in treatment tank. Primary
uncertainty is future influent concentration.
• Irrigation management - Decision rule specifies how current applied irrigation water should
depend on current soil moisture and temperature. Primary uncertainties are future