Date on Master's Thesis/Doctoral Dissertation


Document Type

Doctoral Dissertation

Degree Name

Ph. D.


Industrial Engineering

Committee Chair

Usher, John Stephen

Author's Keywords

Preventive maintenance; Replacement scheduling; Optimization; Maintenance; Scheduling


Plant maintenance--Mathematical models; Scheduling


Preventive maintenance is a broad term that encompasses a set of activities aimed at improving the overall reliability and availability of a system. Preventive maintenance involves a basic trade-off between the costs of conducting maintenance/replacement activities and the cost savings achieved by reducing the overall rate of occurrence of system failures. Designers of preventive maintenance schedules must weigh these individual costs in an attempt to minimize the overall cost of system operation. They may also be interested in maximizing the system reliability, subject to some sort of budget constraint. In this dissertation, we present a complete discussion about the problem definition and review the literature. We develop new nonlinear mixed-integer optimization models, solve them by standard nonlinear optimization algorithms, and analyze their computational results. In addition, we extend the optimization models by considering engineering economy features and reformulate them as a multi-objective optimization model. We optimize this model by generational and steady state genetic algorithms as well as by a simulated annealing algorithm and demonstrate the computational results obtained by implementation of these algorithms. We perform a sensitivity analysis on the parameters of the optimization models and present a comparison between exact and metaheuristic algorithms in terms of computational efficiency and accuracy. Finally, we present a new mathematical function to model age reduction and improvement factor parameter used in optimization models. In addition, we develop a practical procedure to estimate the effect of maintenance activity on failure rate and effective age of multi component systems.