The SMKP set consists of 30 instances of a stochastic multiple knapsack problem. The first-stage problem has 240 binary variables and 50 knapsack constraints. Second-stage problems have 120 binary variables and 5 knapsack constraints. Each instance has 20 scenarios and only the second-stage cost vector is random. The instances are available in smps format here. Details regarding formulation, data generation, and computational results can be found in [1]
REFERENCES:
[1] Gustavo Angulo, Shabbir Ahmed, and Santanu S. Dey. Improving the integer L-shaped method. Submitted, 2014.