In the automotive industry, decentralized in-house logistic areas called supermarkets are used for parts
feeding to mixed-model assembly lines. In the literature, generally, the number of supermarkets and their
locations are determined according to a previously balanced assembly line. As a result, supermarket locations have to be completely dependent on the line balance. There may be more than one solution that
gives the same performance criterion value in the line balancing. Among these alternative solutions, it is
more reasonable to use a solution that provides lower logistics costs than others. For this purpose, in this
study, the mixed-model assembly line balancing problem and the supermarket location problem are considered simultaneously. Accordingly, in order to minimize the total costs of the assembly line and supermarkets, a mixed-integer mathematical model is developed, and the developed model is solved by using
constraint programming (CP). In addition, a new approach based on Ant Colony Optimization (ACO) and
Simulated Annealing (SA) is presented for large-sized problems. An illustrative example problem is solved
using both the developed model and the proposed algorithm. The effectiveness of the proposed approach is
tested through a new data set of test problems presented to the literature for a computational experiment.
The experimental results show that the total costs are reduced meaningfully and a more realistic and applicable structure is achieved by the proposed approach.