Preventive maintenance optimization using a Hybrid Multi-Objective Evolutionary Algorithm

Aitor Goti and Ana Isabel Sanchez
Submission Type: 
Full Paper
phmec_16_063.pdf561.91 KBJune 10, 2016 - 12:45am

This paper is focused on the problem of preventive maintenance optimization in a manufacturing environment, to determine the optimal preventive maintenance frequencies for equipment under cost and profit criteria, considering production, quality and maintenance aspects. The paper is based on a previously developed maintenance model, to execute a benefit and cost optimization process using a Hybrid Multi-Objective Evolutionary Algorithm (Hybrid MOEA). The hybrid algorithm combines a global search method complemented with a local search one. The hybridization is done according to two different schemes. Firstly, ‘a posteriori’ scheme has been implemented, where the MOEA runs for a number of generations obtaining an approximation of the Pareto front to apply then a local search from each non-dominated solution of the front. Secondly, an ‘on-line’ scheme has been developed, where in each generation (or after a reduced number of generations) of the evolutionary algorithm a local search is applied on each non-dominated solution to return then the improved solutions to the MOEA as the current population. Both hybrid schemes have been applied to an industrial manufacturing case where the benefit of implementing the hybrid optimization approach is shown, by comparing the hybrid schemes with the MOEA.

Publication Year: 
Publication Volume: 
Publication Control Number: 
Page Count: 
Submission Keywords: 
hybrid algorithms
preventive maintenance
multi-objective optimization
Submission Topic Areas: 
Industrial applications
Submitted by: 

follow us

PHM Society on Facebook Follow PHM Society on Twitter PHM Society on LinkedIn PHM Society RSS News Feed