Basics of genetic algorithms optimization for RAMS applications

https://doi.org/10.1016/j.ress.2005.11.046Get rights and content

Abstract

This paper discusses the use of genetic algorithms (GA) within the area of reliability, availability, maintainability and safety (RAMS) optimization. First, the multi-objective optimization problem is formulated in general terms and two alternative approaches to its solution are illustrated. Then, the theory behind the operation of GA is presented. The steps of the algorithm are sketched to some details for both the traditional breeding procedure as well as for more sophisticated breeding procedures. The necessity of affine transforming the fitness function, object of the optimization, is discussed in detail, together with the transformation itself. In addition, how to handle constraints by the penalization approach is illustrated. Finally, specific metrics for measuring the performance of a genetic algorithm are introduced.

Section snippets

Making design and maintenance decisions informed on RAMS&C

System reliability and availability optimization is classically based on quantifying the effects that design choices and testing and maintenance activities have on reliability, availability, maintainability (RAM) attributes [1]. A quantitative model is used to asses how the design and maintenance choices affect the system RAM attributes and the involved costs (C). Thus, the design and maintenance optimization problem must be framed as a multiple criteria decision making (MCDM) problem where

Formulation of the multiple objective optimization problem

The commonly accepted manner to tackle the previously illustrated MCDM problem is to formulate it as a multi-objective optimization problem (MOP). A general MOP considers a set of decision variables x, a set of objective functions f(x) and a set of constraints g(x). Adapting the formal definition of Refs. [18], [19], in the RAMS field the MOP regards the optimization of the vector of multi-objective functionsy=f(x)=(R(x),U(x),M(x),Risk(x),C(x))subject to the vector of constraintsg(x)=(R(x)⩾RL,U(

MOP solution approaches

There exist many works in the scientific literature devoted to the solution of the above MOP using different optimization techniques. Two types of approaches are here briefly summarized [21].

In general, an MOP admits multiple solutions due to the conflicting nature of its attributes. Therefore, to arrive at a final decision, the decision maker must make a value judgment among the identified options, giving preference to some attributes at the expense of other, possibly conflicting, ones. The

Genetic algorithms for RAMS optimization

This section introduces the fundamentals of genetic algorithms (GA). The presentation is first made with respect to SOPs and then extended to MOPs.

Discussion

The goal of this paper was to provide an overview of the use of GA for the solution of optimization problems with particular reference to RAMS.

The basics of the methodology have been discussed with respect to RAMS analysis, particularly in the area of redundancy allocation and maintenance and surveillance optimization. The presentation of the material is by no means exhaustive and the references cited do not cover all the work done in the field.

The basic procedures underpinning the functioning

Acknowledegments

The authors acknowledge the benefits received by the critical review of the anonymous referees. Also, the authors wish to thank Maurizio Cipollone for his contribution to the calculations concerning the verification of the efficiency of breading procedures, performed during his thesis project at the Polytechnic of Milan. Finally, the work was performed within the EU-sponsored Thematic Network SAFERELNET, under project number GTC2-2000-33043.

References (31)

  • G. Levitin et al.

    Join redundancy and maintenance optimization for multistage series-parallel systems

    Rel Eng Syst Safety

    (1999)
  • S. Martorell et al.

    Constrained optimization of test intervals using a steady-state genetic algorithm

    Reliab Eng Sys Saf

    (2000)
  • S. Martorell et al.

    Alternatives and challenges in optimizing industrial safety using genetic algorithms

    Reliab Eng Sys Saf

    (2004)
  • Muñoz A, Martorell S, Serradell V. Numerical absolute and constrained optimization of maintenance based on risk and...
  • M. Harunuzzaman et al.

    Optimization of standby safety system maintenance schedules in nuclear power plants

    Nucl Technol

    (1996)

    • Cited by (0)

    View full text
    Copyright © 2005 Elsevier Ltd. All rights reserved.