STOCHASTIC LINEAR-PROGRAMS WITH RESTRICTED RECOURSE

Stochastic programs with recourse provide an effective modelling parad igm for sequential decision problems with uncertain or noisy data, whe n uncertainty can be modelled by a discrete set of scenarios. In two-s tage problems the decision variables are partitioned into two groups: a set of structural, first-stage decisions, and a set of second-stage, recourse decisions. The structural decisions are scenario-invariant, but the recourse decisions ate scenario-dependent and can vary substan tially across scenarios. In several applications it is important to re strict the variability of recourse decisions across scenarios, or to i nvestigate the tradeoffs between the stability of recourse decisions a nd expected cost of a solution. We present formulations of stochastic programs with restricted recourse that trade off recourse stability wi th expected cost. The models generate a sequence of solutions to which recourse robustness is progressively enforced via parameterized, sati sficing constraints. We investigate the behavior of the models on seve ral test cases, and examine the performance of solution procedures bas ed on the primal-dual interior point method. (C) 1997 Elsevier Science B.V.