Multi-objective Cooperative Co-evolutionary Algorithm Research Articles

Network planning of metro-based underground logistics system against mixed uncertainties: A multi-objective cooperative co-evolutionary optimization approach

Featured with zero-carbon, jam-free, and high-capacity, the utilization of metro systems for collaborative passenger-and-freight transport (i.e., the metro-based underground logistics system, M-ULS) has been recognized as a favorable alternative to realize automated goods movement in future megacities. This paper proposes an equilibrium chance-constrained programming (ECP) approach for the M-ULS network planning problem (MULNP) considering uncertain demand and costs. Based on the ongoing satellite city development project, the layout formation, facility components, and transportation workflows of a two-tier M-ULS network are designed. The joint decisions (i.e., location, allocation, capacity, inventory, spoke pipeline layout, and flow routing) are formulated as an ECP model with conjoint objectives of minimal fixed cost, operating cost, and penalty cost due to low operating load, wherein the chance-based constraints are further reformulated into crisp equivalent form. To yield high-quality solutions for MULNP problem, an improved multi-objective cooperative co-evolutionary algorithm (MoCC) incorporating non-dominated sorting and chromosomal recombination strategy is proposed and then validated via two groups of instances. Results from the Beijing case study suggest that the proposed method has computational advantages over MoPSO and NSGA-II in acquiring Pareto fronts of large-scale MULNP. The best configurations and layout schemes of the M-ULS networks in both uncertain and deterministic scenarios are analyzed. Based on the findings, several insights for real-world M-ULS network decision-making are provided.

A scalable parallel cooperative coevolutionary PSO algorithm for multi-objective optimization

We present a parallel multi-objective cooperative coevolutionary variant of the Speed-constrained Multi-objective Particle Swarm Optimization (SMPSO) algorithm. The algorithm, called CCSMPSO, is the first multi-objective cooperative coevolutionary algorithm based on PSO in the literature. SMPSO adopts a strategy for limiting the velocity of the particles that prevents them from having erratic movements. This characteristic provides the algorithm with a high degree of reliability. In order to demonstrate the effectiveness of CCSMPSO, we compare our work with the original SMPSO and three different state-of-the-art multi-objective CC metaheuristics, namely CCNSGA-II, CCSPEA2 and CCMOCell, along with their original sequential counterparts. Our experiments indicate that our proposed solution, CCSMPSO, offers significant computational speedups, a higher convergence speed and better or comparable results in terms of solution quality, when evaluated against three other CC algorithms and four state-of-the-art optimizers (namely SMPSO, NSGA-II, SPEA2, and MOCell), respectively. We then provide a scalability analysis, which consists of two studies. First, we analyze how the algorithms scale when varying the problem size, i.e., the number of variables. Second, we analyze their scalability in terms of parallelization, i.e., the impact of using more computational cores on the quality of solutions and on the execution time of the algorithms. Three different criteria are used for making the comparisons, namely the quality of the resulting approximation sets, average computational time and the convergence speed to the Pareto front.

Multi-objective co-operative co-evolutionary algorithm for minimizing carbon footprint and maximizing line efficiency in robotic assembly line systems

Methods for reducing the carbon footprint is receiving increasing attention from industry as they work to create sustainable products. Assembly line systems are widely utilized to assemble different types of products and in recent years, robots have become extensively utilized, replacing manual labor. This paper focuses on minimizing the carbon footprint for robotic assembly line systems, a topic that has received limited attention in academia. This paper is primarily focused on developing a mathematical model to simultaneously minimize the total carbon footprint and maximize the efficiency of robotic assembly line systems. Due to the NP-hard nature of the considered problem, a multi-objective co-operative co-evolutionary (MOCC) algorithm is developed to solve it. Several improvements are applied to enhance the performance of the MOCC for obtaining a strong local search capacity and faster search speed. The performance of the proposed MOCC algorithm is compared with three other high-performing multi-objective methods. Computational and statistical results from the set of benchmark problems show that the proposed model can reduce the carbon footprint effectively. The proposed MOCC outperforms the other three methods by a significant margin as shown by utilizing one graphical and two quantitative Pareto compliant indicators.

A Multiobjective Cooperative Coevolutionary Algorithm for Hyperspectral Sparse Unmixing

Sparse unmixing of hyperspectral data is an important technique aiming at estimating the fractional abundances of the endmembers. Traditional sparse unmixing is faced with the $\boldsymbol {l_{0}}$ -norm problem which is an NP-hard problem. Sparse unmixing is inherently a multiobjective optimization problem. Most of the recent works combine cost functions into single one to construct an aggregate objective function, which involves weighted parameters that are sensitive to different data sets and difficult to tune. In this paper, a novel multiobjective cooperative coevolutionary algorithm is proposed to optimize the reconstruction term, the sparsity term and the total variation regularization term simultaneously. A problem-dependent cooperative coevolutionary strategy is designed because sparse unmixing encounters a large scale optimization problem. The proposed approach optimizes the nonconvex $\boldsymbol {l_{0}}$ -norm problem directly and can find a better compromise between two or more competing cost function terms automatically. Experimental results on simulated and real hyperspectral data sets demonstrate the effectiveness of the proposed method.

Multi-objective constructive cooperative coevolutionary optimization of robotic press-line tending

ABSTRACTThis article investigates multi-objective optimization of the robot trajectories and position-based operation-coordination of complex multi-robot systems, such as press lines, to improve the production rate and obtaining smooth motions to avoid excessive wear of the robots' components. Different functions for handling the multiple objectives are evaluated on real-world press lines, including both scalarizing single-objective functions and Pareto-based multi-objective functions. Additionally, the Multi-Objective Constructive Cooperative Coevolutionary () algorithm is proposed, for Pareto-based optimization, which uses a novel constructive initialization of the subpopulations in a co-adaptive fashion. It was found that Pareto-based optimization performs better than the scalarizing single-objective functions. Furthermore, gives substantially better results compared to manual online tuning, as currently used in the industry. Optimizing robot trajectories and operation-coordination of complex multi-robot systems using the proposed method with significantly improves productivity and reduces maintenance. This article hereby addresses the lack of systematic methods for effectively improving the productivity of press lines.

MOCCA-II: A multi-objective co-operative co-evolutionary algorithm

Most real-world problems naturally involve multiple conflicting objectives, such as in the case of attempting to maximize both efficiency and safety of a working environment. The aim of multi-objective optimization algorithms is to find those solutions that optimize several components of a vector of objective functions simultaneously. However, when objectives conflict with each other, the multi-objective problem does not have a single optimal solution for all objectives simultaneously. Instead, algorithms attempt to search for the set of efficient solutions, known as the global non-dominated set, that provides solutions that optimally trade-off the objectives. The final solution to be adopted from this set would depend on the preferences of the decision-makers involved in the process. Hence, a decision-maker is typically interested in knowing as many potential solutions as possible.In this paper, we propose an extension to a previous piece of work on multi-objective cooperative coevolution algorithms (MOCCA). The idea was motivated with a practical problem in air traffic management to design terminal airspaces. MOCCA and a further study that was done on a distributed environment for MOCCA, were found to fit the need for the application. We systematically questioned key components of this algorithm and investigated variations to identify a better design. This paper summarizes this systematic investigation and present the resultant new algorithm: multi-objective co-operative co-evolutionary algorithm II (MOCCA-II).

Multi-Objective Cooperative Co-Evolutionary Algorithm for Negotiated Scheduling of Distribution Supply Chain

PDF HTML阅读 XML下载 导出引用 引用提醒 分销供应链协商调度多目标合作协同演化算法 DOI: 10.3724/SP.J.1001.2013.04288 作者: 作者单位: 作者简介: 通讯作者: 中图分类号: 基金项目: 国家自然科学基金(60904072); 国家教育部博士点新教师基金(20090185120002); 国家教育部人文社科青年基金(09YJC630018); 电子科技大学中央高校基本科研业务经费(103.1.2 E022050205) Multi-Objective Cooperative Co-Evolutionary Algorithm for Negotiated Scheduling of Distribution Supply Chain Author: Affiliation: Fund Project: 摘要 | 图/表 | 访问统计 | 参考文献 | 相似文献 | 引证文献 | 资源附件 | 文章评论 摘要:研究了在制造商占优并优先调度的分销供应链中,多个分销商同时与制造商进行协商以改善自身调度的问题,建立了基于补偿的多目标协商调度模型,提出了同时实施分销商局部演化计算与制造商全局演化计算的新型多目标合作协同演化算法GLCCEC.提出了制造商全局精英解的跳跃渐变解组合策略及全局非支配解集实时更新策略,设计了保持局部作业顺序约束下的分销商局部解全局化动态规划算法.实验结果表明,GLCCEC算法能够在不损害制造商调度的条件下有效改善每个分销商的调度,所获得的非支配解集不仅目标值优于现有的3 种主要合作协同演化算法MOCCGA,NSCCGA,GBCCGA,而且具有良好的解分散度. Abstract:It is investigated that multiple distributors simultaneously negotiate with a manufacturer to improve themselves schedules on a distribution supply chain in which manufacturer has stronger power than distributors and does scheduling decision prior to distributors. A compensation based negotiation scheduling model is built. A novel multi-objective cooperative co-evolutionary algorithm (GLCCEC) that concurrently implements local evolutionary computing of distributors and global evolutionary computing of manufacturer is proposed. Global elite solution combination strategy with gradually gene skipping change and real time updating of global non-dominated solution set are designed for manufacturer. A dynamic programming algorithm with constraint of retaining sequence of local schedule is designed in order to get global solution from a local solution of distributor. Computational experiments show that GLCCEC algorithm can effectively improve schedule of each distributor with no deterioration of manufacturer’s schedule. Moreover, the non-dominated solutions of GLCCEC not only are better than that of other best cooperative co-evolutionary algorithms: MOCCGA, NSCCGA, GBCCGA, but also has good spread in solution space. 参考文献 相似文献 引证文献

On generating interpretable and precise fuzzy systems based on Pareto multi-objective cooperative co-evolutionary algorithm

A novel approach to construct a set of interpretable and precise fuzzy systems based on the Pareto multi-objective cooperative co-evolutionary algorithm (PMOCCA) is proposed in this paper. Firstly, feature selection is used to reduce the dimensionality of the data in order to improve the performance and reduce computational burden. Secondly, the fuzzy clustering algorithm is employed to identify the initial fuzzy system. Thirdly, the PMOCCA is carried out to evolve the initial fuzzy system to optimize the number of rules, the antecedents of the rules and the parameters of the antecedents simultaneously. Furthermore, the interpretability-driven simplification techniques are used iteratively to reduce the fuzzy systems, thus the interpretability of the fuzzy systems is improved. Finally, the proposed approach is applied to benchmark problems, and the results show its validity.