An ant colony optimisation approach considering jointly scheduling and preventive maintenance in the flowshop sequencing problem

This paper presents INTACO, a hybrid Ant Colony Optimisation (ACO) algorithm coupled with a local search applied to the joint production and preventive maintenance scheduling problem in the flowshop sequencing problem. INTACO uses pheromone trail information to perform modifications on complete joint production and preventive maintenance solutions unlike more traditional ant systems that use pheromone trail information to construct complete solutions. Several new interesting features are proposed and evaluated. In particular, the use of a common representation of preventive maintenance and production data to optimise a common objective function which takes into account both preventive maintenance and production criteria with a new pheromone evaluation rule. Moreover, to enhance the performances of the proposed ACO algorithm, new local search procedures for ants are proposed. INTACO is tested on a set of non-standard test problems. We compare the results obtained to those of a genetic algorithm developed in previous works. The results and experiments carried out indicate that the proposed ant-colony algorithm provide very effective solutions for this problem.