Abstract
Many real world supply chains involve uncertain order lead times which impacts order policies. The Robust Optimization paradigm has been established as a modeling approach for problems that involve such uncertainties and is tractable for several classes of problems. We present a robust optimization model of a multi-period inventory control problem for a retailer under lead time uncertainty. A polyhedral lead time uncertainty set is devised by parameterizing the lead time of orders at each time period. We show that this problem is an application of robust optimization under column-wise uncertainty, with uncertainty characterized by the coefficients for orders at each time period. As a result, the standard robust optimization reformulation technique utilizing duality cannot be used for these problems. Instead, a variation of Benders' decomposition is used to determine optimal robust (i.e., best worst-case) policy parameters under two policies (static and basestock). The computational performance of the algorithm is particularly good for the basestock policy, capable of solving problem instances with 100 time periods in thirty seconds.
Original language | English (US) |
---|---|
Pages | 3869-3878 |
Number of pages | 10 |
State | Published - 2013 |
Event | IIE Annual Conference and Expo 2013 - San Juan, Puerto Rico Duration: May 18 2013 → May 22 2013 |
Other
Other | IIE Annual Conference and Expo 2013 |
---|---|
Country/Territory | Puerto Rico |
City | San Juan |
Period | 5/18/13 → 5/22/13 |
All Science Journal Classification (ASJC) codes
- Industrial and Manufacturing Engineering