Elsevier

Applied Soft Computing

Volume 71, October 2018, Pages 553-567
Applied Soft Computing

A hybrid multi-objective AIS-based algorithm applied to simulation-based optimization of material handling system

https://doi.org/10.1016/j.asoc.2018.07.034Get rights and content

Highlights

  • A hybrid multi-objective optimization algorithm derived from immunological and biological evolution concepts is proposed.

  • An optimization approach integrating the proposed algorithm with an industrial-grade simulator is implemented.

  • The optimization approach’s performance is examined via numerical benchmark problems and a real-life simulation study.

  • The results reveal its ability of helping management to find near optimal system operating conditions and parameters.

Abstract

Optimization of complex real-world problems often involves multiple objectives to be considered simultaneously. These objectives are often non-commensurable and competing. For example, material handling is a vital element of industrial processes, which involves a variety of operations including the movement, storage and control of materials throughout the processes of manufacturing, distribution, and disposal while having to satisfy multiple objectives. Having an efficient and effective material handling system (MHS) is of great importance to various industries, such as manufacturing and logistics industries, for maintaining and facilitating a continuous flow of materials through the workplace and guaranteeing that required materials are available when needed. In this paper, a hybrid multi-objective optimization algorithm largely based on Artificial Immune Systems (AIS) is applied to simulation-based optimization of material handling system. This proposed algorithm hybridizes the AIS with the Genetic Algorithm (GA) by incorporating the crossover operator derived from the biological evolution. The reason behind such hybridization is to further enhance the diversity of the clone population and the convergence of the algorithm. In this paper, other than conducting numerical experiments to assess the performance of the proposed algorithm by using several benchmark problems, the proposed algorithm is also applied to optimize a multi-objective simulation-based problem on a material handling system in order to demonstrate the applicability of the proposed algorithm in real-life cases. The results show that for most cases the proposed algorithm outperforms the other benchmark algorithms especially in terms of solution diversity.

Introduction

In real world, many problems, no matter whether they are in the domain of engineering, finance, distribution or manufacturing, etc., can be formulated into different forms of optimization problems. Most of these real-world optimization problems normally involve multiple objectives rather than one single objective. Solving this kind of problems is never an easy task because objectives of such problems are often found to be at least partly non-commensurable and conflicting. Very often, there is no single best solution to the multi-objectives optimization problems, but rather a set of optimal trade-off solutions that exists among the objectives. During the solution evaluation process, a huge number of alternative solutions are required to be evaluated. However, it is very difficult to evaluate all possible solutions as this requires tremendous computer resources that normally are not available. Therefore, an efficient and effective optimization algorithm is needed to guide the search process to the promising areas of the solution space and eventually the optimal solutions, which can aid the decision makers in arriving at an optimal solution efficiently and effectively.

Competitive market factors, such as more stringent government regulations, fierce competition, and demanding customers, in recent years have created a tremendous pressure on the management in all supply chain parties including distributors and manufacturers. This has become a pressing need to improve the efficiency and effectiveness of their distribution and manufacturing systems. To this end, the ability of analyzing and evaluating these systems and related operations involving the deployment of complex material handling systems, are vital for distribution and manufacturing practitioners. Such problems are traditionally investigated by analytical techniques such as linear programming. However, due to the more complex business environment, the real-world systems become increasingly more complex including complex interactions among system components and unexplained randomness and uncertainties. The analysis of such complex stochastic systems is often a difficult task and makes analytical methods hardly applicable. To overcome these issues, these systems can be studied and optimized more effectively and efficiently by integrating simulation modeling with optimization algorithm because simulation techniques alone can only provide feasible solutions of these complex systems. In order to improve the quality of solutions through simulation, an optimization algorithm to direct the simulation process to attain optimality is essential and academically interesting with great practical value.

In the literature, different metaheuristic algorithms, such as Genetic Algorithm (GA) [1], Evolution Strategy (ES) [2], Simulated Annealing (SA) [3], Differential Evolution (DE) [4], Particle Swarm Optimization (PSO) [5,6], Artificial Immune Systems (AIS) [7], etc., have been developed for solving multi-objective simulation-based optimization problems. However, AIS based on the mechanisms and concepts of biological immune system have received special attention among the research community because the biological immune system provides a rich source of inspiration to the research community with their interesting characteristics: distributed nature, self-organization, memory and learning capabilities. In this paper, we make use of a simulation-based optimization method for solving a multi-objective simulation-based optimization problem associated with material handling system in distribution and manufacturing industries. The optimization method is developed based on the integration of a hybrid multi-objective AIS-based algorithm [8] and a commercial simulation tool (FlexSim). Two AIS theories, namely, clonal section principle and immune network theory are the backbone for building the algorithm, which are supplemented by a crossover operation derived from GA to enhance the performance in diversity and convergence.

The rest of this paper is organized as follows: Section 2 presents the basis for the design of the proposed algorithm and introduces the major features of the proposed algorithm. Sections 3 and 4 assess the performance of the algorithm through a set of numerical optimization experiments and a simulation-based optimization study on a material handling system respectively by benchmarking with some well-known optimization algorithms. Section 5 concludes the work done in this paper and discusses future research directions.

Section snippets

Multi-objective simulation-based optimization

Finding the solutions to the multi-objective optimization problems has long been a challenge to researchers because both the Pareto optimality and the diversity of the generated solutions must be simultaneously addressed. Unlike solutions in single objective optimization problems, which can easily be compared according to the value of the objective function, solutions in multi-objective problems cannot directly be compared with each other unless employing classical techniques, such as, weighted

Numerical experiments

In this benchmark study, a set of experiments based on several multi-objective numerical optimization problems was performed to benchmark the proposed algorithm with other well-known multi-objective optimization algorithms, that is, two immune algorithms – MISA [49] and NNIA [18] and two other evolutionary algorithms – NSGA-II [13] and SPEA2 [14]. All these experiments were conducted using a computer with Xeon E5-2620 2 GHz CPU with 2 GB RAM and the Excel with VBA was used as an implementation

Simulation-based optimization benchmark study

In this benchmark study, a set of experiments based on a real-life multi-objective optimization problem was performed to evaluate the performance and capability of the proposed optimization method. All these experiments were conducted using a computer with Xeon E5-2620 2 GHz CPU with 2 GB RAM.

Conclusions and future work

This research applies a hybrid multi-objective AIS-based algorithm called SCMIA for solving numerical benchmark optimization problems and evaluating the system optimality in the domain of material handling with respect to certain criteria through simulation modeling. The characteristics of the algorithm and the concept of the multi-objective simulation-based optimization are discussed. The performance of the algorithm in solving several numerical benchmark problems are evaluated and analyzed.

References (60)

  • E.Y.C. Wong et al.

    Immunity-based hybrid evolutionary algorithm for multi-objective optimization in global container repositioning

    Eng. Appl. Artif. Intell.

    (2009)
  • F. Javidrad et al.

    A new hybrid particle swarm and simulated annealing stochastic optimization method

    Appl. Soft Comput.

    (2017)
  • C.S.K. Leung et al.

    An optimization framework for modeling and simulation of dynamic systems based on AIS

  • S. García et al.

    Advanced nonparametric tests for multiple comparisons in the design of experiments in computational intelligence and data mining: experimental analysis of power

    Inf. Sci.

    (2010)
  • D. Goldberg

    Genetic Algorithms in Search, Optimization, and Machine Learning

    (1989)
  • I. Rechenberg

    Cybernetic solution path of an experimental problem

    Royal Aircraft Establishment Translation No. 1122, B. F. Toms, Trans., Ministry of Aviation

    (1965)
  • S. Kirkpatrick et al.

    Optimization by simulated annealing

    Science

    (1983)
  • R. Storn et al.

    Differential evolution – a simple and efficient heuristic for global optimization over continuous spaces

    J. Global Optim.

    (1997)
  • J. Kennedy et al.

    Particle swarm optimization

    Proceedings of the IEEE International Conference on Neural Networks

    (1995)
  • R. Eberhart et al.

    A new optimizer using particle swarm theory

    Proceedings of the Sixth International Symposium on Micro Machine and Human Science

    (1995)
  • L.N. de Castro et al.

    Artificial Immune Systems: A New Computational Intelligence Approach

    (2002)
  • C.S.K. Leung et al.

    A hybrid multi-objective immune algorithm for numerical optimization

  • K. Deb

    Multi-Objective Optimization Using Evolutionary Algorithms

    (2001)
  • N.C. Cortés et al.

    Multiobjective optimization using ideas from the clonal selection principle

  • I. Das et al.

    A closer look at drawbacks of minimizing weighted sums of objectives for Pareto set generation in multicriteria optimization problems

    Struct. Multidiscipl. Optim.

    (1997)
  • J.D. Schaffer

    Multiple objective optimization with vector evaluated genetic algorithms

    Proceedings of 1st International Conference on Genetic Algorithms, L

    (1985)
  • K. Deb et al.

    A fast elitist non-dominated sorting genetic algorithm for multi-objective optimisation: NSGA-II

  • E. Zitzler et al.

    SPEA2: improving the strength pareto evolutionary algorithm

  • D.W. Corne et al.

    PESA-II: regionbased selection in evolutionary multiobjective optimization

  • C.A. Coello Coello et al.

    MOPSO: a proposal for multiple objective particle swarm optimization

    Proceedings of the 2002 Congress on Evolutionary Computation

    (2002)
  • Cited by (16)

    • A full-featured cooperative coevolutionary memory-based artificial immune system for dynamic optimization

      2022, Applied Soft Computing
      Citation Excerpt :

      Biological immune system has outstanding qualities, which can help it adapt to various changes in the body. AISs have been utilized in a wide variety of application areas, such as anomaly detection [12], data classification [13], global optimization [14], multi-objective optimization [15], robotics [16], and recommender systems [17]. However, they have rarely exploited in the area of dynamic optimization problems.

    • A Relation B-cell Network used for data identification and fault diagnosis

      2021, Applied Soft Computing
      Citation Excerpt :

      Artificial Immune System (AIS) is a series of intelligent identification methods which are inspired by biological immune systems. Currently, many AIS, such as negative selection [1–3] proposed by Forrest [4], clone selection and immune network [5] have been developed for their interpretability and observability and wildly applied in various fields of computational intelligence, includes optimization [6–11] data clustering [12–15], classification [16–19], anomaly detection [20–23], network intrusion detection [24–28], computer security-related applications [29,30] and so on. However, these methods are still machine learning algorithms based on hand-crafted features, and the final task accuracy is often largely affected by the quality of artificial features [31].

    • A multi-objective immune algorithm with dynamic population strategy

      2019, Swarm and Evolutionary Computation
      Citation Excerpt :

      At last, the conclusions of this paper are summarized in Section 5. Artificial immune system is a kind of heuristic algorithms imitating the information processing mechanism of biological immune system, which has found numbers of applications in handling material in industries [38], solving fault detection and isolation problem [39,40] and character recognition system [41]. Especially immune algorithm has been successfully applied for MOPs and shown pretty promising performance in solving 3D terrain delayment of wireless sensor network [42], nonlinear interval-valued programming [43], design recommendation system [44] and scheduling problems [45].

    View all citing articles on Scopus
    View full text