A comparison of logic and mixed-integer programming solvers for batch sequencing with sequence-dependent setups

A batch sequencing model with sequence-dependent setup-times and -costs is used to compare modelling and solving with two different general solvers. "Conceptual models" are implemented in the constraint propagation / logic programming language CHARME and solved with the PROLOG inference engine. The mixed-integer-programming (MIP) formulation of the same problem is solved with OSL, a state-of-the-art MIP solver. Modelling is easier in CHARME and computational results show, that the first approach outperforms the second one for instances with high capacity utilization.