A new Lagrangian Relaxation algorithm for single-stage chemical scheduling

A Lagrangian relaxation-based solving framework for the scheduling of single-stage chemical process is presented. The problem is formulated as an Integer Programming model. Unlike the regular procedure of LR algorithm, the relaxed problem can be optimally solved without being decomposed into a set of smaller subproblems. Therefore, the computational effort can be dramatically reduced. By testing 50 types of random problem instances, this solution strategy is found to be effective, in terms of both quality of solution and computational requirements.