Analytical investigations in process synthesis and process operations
Two distinct solution philosophies have emerged through the study of process synthesis and process operations problems in recent years. Heuristics offer fast solutions but no guarantee of optimality. Optimization approaches, on the other hand, offer rigor but suffer combinatorial explosion of computational requirements.
Our work in this area addresses the following fundamental questions. Is it possible to introduce rigor in the use and practice of heuristics? Is it possible to solve combinatorial optimization problems in an approximate way which escapes the curse of dimensionality without a significant compromise of optimality?
We pursue analytical investigations to theoretically determine the computational complexity of process synthesis, planning and scheduling problems, analyze the worst-case and expected behavior of existing heuristics, and develop new, optimization-based, polynomial-time heuristics that are optimal in expectation.
The long term goal of this work is to build an optimization-based science of heuristics that will lead to the design of new, provably optimal heuristics for process synthesis and operations.
- Furman, K. C. and N. V. Sahinidis, Approximation algorithms for the minimum number of matches problem in HENS, Industrial & Engineering Chemistry Research, 43(14), 3554-3565, 2004.
- Ahmed, S. and N. V. Sahinidis, An approximation scheme for stochastic integer programs arising in capacity expansion, Operations Research, 51(3), 461-471, 2003.
- K. C. Furman and N. V. Sahinidis, Computational Complexity of Heat Exchanger Network Synthesis, Computers & Chemical Engineering, 25(9-10), 1371-1390, 2001.
- Ahmed, S. and N. V. Sahinidis, Analytical Investigations of the Process Planning Problem, Computers & Chemical Engineering, 23(11-12), 1605-1621, 2000.
- Liu, M. L. and N. V. Sahinidis, Bridging the Gap Between Heuristics and Optimization: The Capacity Expansion Case, AIChE Journal, 43(9), 2289-2299, 1997.