Traditional optimization approaches focus on iteratively searching the design universe and converging to the best one. However, these approaches can be time consuming. Even the simulation of a single design can be expensive because accurately estimating performance measures usually requires long simulation. Thus, finding the best design is often infeasible for large discrete-event systems.

Instead of insisting on picking the best design, Ordinal Optimization concentrates on finding good, or better, designs and reduces the required simulation time dramatically. Ordinal Optimization has been applied on a 10-node network, where it is shown that we can isolate a good design with high probability with relatively short simulations, instead of long simulations. They also demonstrate many orders of magnitude of speedup.

A crucial issue to apply Ordinal Optimization is the ability of knowing when we are confident or satisfied enough with the ordinal results, e.g., a good subset has been determined with high probability. To fully utilize the advantages of ordinal Optimization, in this subject, we are developing effective approaches to quantify the simulation confidence level, particularly for large discrete-event systems.

A highly recommended wet site for ordinal optimization is the one developed by the inventor, Professor Yu-Chi (Larry) Ho at Harvard University.

Selected Publications

• "Scheduling Semiconductor Wafer Fabrication by Using Ordinal Optimization-Based Simulation," with B. W. Hsieh and S. C. Chang, IEEE Transactions on Robotics and Automation, Vol. 17, No. 5, pp. 599-608, October 2001.
• "Large Convergence Properties of Two-Stage Stochastic Programming," with L. Dai and J. R. Birge, Journal of Optimization Theory and Applications, September 2000.
• "Simulation Budget Allocation for Further Enhancing the Efficiency of Ordinal Optimization," with J. Lin, E. Yücesan, and S. E. Chick, Journal of Discrete Event Dynamic Systems: Theory and Applications, July 2000.
• "Computing Efforts Allocation for Ordinal Optimization and Discrete Event Simulation," with H. C. Chen, and E. Yucesan, IEEE Transactions on Automatic Control, May 2000.
• "Ordinal Optimization and Simulation," with Y. C. Ho, C. G. Cassandras, and L. Dai, Journal of Operational Research Society, April 2000.
• "A Hybrid Approach for Integer Programming Combining Genetic Algorithms, Linear Programming and Ordinal Optimization," with Y. C. Luo and M. Guignard-Spielberg, To appear in Journal of Intelligent Manufacturing.
• "Distributed Web-Based Simulation Optimization," with Y. C. Luo, C. H. Chen, E. Yücesan, and I. Lee, Proceedings of the 2000 Winter Simulation Conference, December 2000.
• "A Hybrid Approach for Integer Programming Combining Genetic Algorithms, Linear Programming and Ordinal Optimization," with Y. C. Luo, and M. Guignard-Spielberg, Proceedings of the Latin-Ibero-American Conference on Operations Research and Systems, September 2000.
• "Stochastic Discrete Resource Allocation Optimization," with L. Shi, Proceedings of 38th IEEE Conference on Decision and Control, December 1999.
• "An Asymptotic Allocation for Simultaneous Simulation Experiments," with H. C. Chen, E. Yücesan, and J. Lin, Proceedings of the 1999 Winter Simulation Conference, December 1999.
• "Fast Fab Scheduling Rule Selection by Ordinal Comparison-Based Simulation," with B. W. Hsieh and S. C. Chang, Proceedings of The International Symposium on Semiconductor Manufacturing, October 1999.
• "Exponential Convergence of Two-Stage Stochastic Programming," Proceedings of The 14th World Congress of International Federation of Automatic Control, with L. Dai and J. R. Birge, July 1999.
• "On the Convergence Rate of Ordinal Optimization for A Class of Stochastic Discrete Resource Allocation," with L. Dai, and J. R. Birge, Proceedings 37th IEEE Conference on Decision and Control, 1998.
• "Dispatching Rule Selection for Semiconductor Wafer Fabrication by Ordinal Optimization and Simulation," with B. W. Hsieh, S. C. Chang, Proceedings of the Rensselaer's Sixth International Conference on Agile, Intelligent, and Computer Integrated Manufacturing, 1998.
• "On The Role of Large Deviation Principle in Ordinal Comparison for Discrete Event Dynamic Systems," with L. Dai, Proceedings of the 36th IEEE Conference on Decision and Control, December 1997.
• ""Comparison of Performance Orders of Different Designs for Discrete-Event Dynamic System," with L. Dai, Proceedings of the 1997 Intelligent Systems and Semiotics Conference, September 1997.
• "Rate of Convergence for Ordinal Comparison of Dependent Simulations in Discrete Event Dynamic Systems," with L. Dai, Journal of Optimization Theory and Applications, July 1997.
• "The Large Deviation principle and Ordinal Comparison in Discrete Event Dynamic Systems," with L. Dai, Proceedings of The Second Chinese World Congress on Intelligent Control and Intelligent Automation, June 1997.
• "SIMD Parallel Discrete Event Dynamic System Simulation," with N. Patsis and M. Larson, IEEE Transactions on Control Systems Technology, January 1997.
• "A Lower Bound for the Correct Subset-Selection Probability and Its Application to Discrete Event System Simulations," IEEE Transactions on Automatic Control, August 1996.
• "The Effect of Correlation on Ordinal Comparison of Discrete Event Dynamic Systems," with L. Dai, Proceedings of the 35th IEEE Conference on Decision and Control, December 1996.

Back to Professor Chun-Hung Chen's Page