Facility location under uncertain environments is an important and challenging problem. The problem deals with the optimal placement of facilities that serve a set of spatially distributed nodes. One way to deal with ...
详细信息
Facility location under uncertain environments is an important and challenging problem. The problem deals with the optimal placement of facilities that serve a set of spatially distributed nodes. One way to deal with this problem is to model uncertainty by means of scenarios and to optimise some robustness criteria such as the average and maximum regrets over these scenarios. We propose to model the robust design as a bi-objective optimisation problem and to use a well-known multi-objective evolutionary algorithm, the NSGA-II, to solve it. We also propose to use the bi-objective optimisation framework to analyse the effects of variations in the number of facilities to install, and of nodes to be served, on the quality of the Pareto solutions. Computational experiments show that the proposal can be used to design robust solutions and to study the effects of changes in the system parameters on the quality of the generated solutions.
Cloud services are required to be composed as a single service to fulfill the workflow applications. Service composition in Cloud raises new challenges caused by the diversity of users with different QoS requirements ...
详细信息
Cloud services are required to be composed as a single service to fulfill the workflow applications. Service composition in Cloud raises new challenges caused by the diversity of users with different QoS requirements and vague preferences, as well as the development of cloud computing having geographically distributed characteristics. So the selection of the best service composition is a complex problem and it faces trade-off among various QoS criteria. In this paper, we propose a Cloud service composition approach based on evolutionaryalgorithms, i.e., NSGA-II and MOPSO. We utilize the combination of multi-objectiveevolutionary approaches and Decision-Making method (AHP) to solve Cloud service composition optimization problem. The weights generated from AHP are applied to the Crowding Distance calculations of the above two evolutionaryalgorithms. Our algorithm beats single-objectivealgorithms on the optimization ability. And compared with general multi-objectivealgorithms, it is able to precisely capture the users' preferences. The results of the simulation also show that our approach can achieve a better scalability.
A computational intelligence approach to system-of-systems architecting is developed using multi-objective optimization. Such an approach yields a set of optimal solutions (the Pareto set) which has both advantages an...
详细信息
A computational intelligence approach to system-of-systems architecting is developed using multi-objective optimization. Such an approach yields a set of optimal solutions (the Pareto set) which has both advantages and disadvantages. The primary benefit is that a set of solutions provides a picture of the optimal solution space that a single solution cannot. The primary difficulty is making use of a potentially infinite set of solutions. Therefore, a significant part of this approach is the development of a method to model the solution set with a finite number of points allowing the architect to intelligently choose a subset of optimal solutions based on criteria outside of the given objectives. The approach developed incorporates a meta-architecture, multi-objective genetic algorithm, and a corner search to identify points useful for modeling the solution space. This approach is then applied to a network centric warfare problem seeking the optimum selection of twenty systems. Finally, using the same problem, it is compared to a hybrid approach using single-objective optimization with a fuzzy logic assessor to demonstrate the advantage of multi-objective optimization.
Determining the contribution of an agent to a system-level objective function (credit assignment) is a key area of research in cooperative multiagent systems. multi-objective optimization is a growing area of research...
详细信息
ISBN:
(纸本)9781450337717
Determining the contribution of an agent to a system-level objective function (credit assignment) is a key area of research in cooperative multiagent systems. multi-objective optimization is a growing area of research, though mostly focused on single agent settings. Many real-world problems are multiagent and multi-objective, (e.g., air traffic management, scheduling observations across multiple exploration robots) yet there is little work on their intersection. In this work, we leverage recent advances in single-objectivemultiagent learning to address multi-objective domains. We focus on the impact of difference evaluation functions (which extracts an agent's contribution to the team objective) on the Non-dominated Sorting Genetic algorithm-II (NSGA-II), a state-of-the-art multi-objective evolutionary algorithm. We derive multiple methods for incorporating difference evaluations into the NSGA-II framework, and test each in a multiagent rover exploration domain, which is a good surrogate for a wide variety of distributed scheduling and resource gathering problems. We show that how and where difference evaluations are incorporated in the NSGA-II algorithm is critical, and can either provide significant benefits or destroy system performance, depending on how it is used. Median performance of the correctly used difference evaluations dominates best-case performance of NSGA-II in a multiagent multi-objective problem.
In this paper, a new proportion entropy function is proposed as an objective function to obtain a well-diversified portfolio. Secondly, a new fuzzy multi-objective portfolio selection model based on the proposed entro...
详细信息
In this paper, a new proportion entropy function is proposed as an objective function to obtain a well-diversified portfolio. Secondly, a new fuzzy multi-objective portfolio selection model based on the proposed entropy function is presented. By using the model, we can find tradeoffs between risk, return and the diversification degree of portfolio. Thirdly, a new multi-objective evolutionary algorithm is designed to solve the proposed model. Finally, some numerical examples are presented to illustrate the effectiveness of the proposed model and the corresponding algorithm.
In this paper, the comparison of multi-objective evolutionary algorithm (MOEA) and Single-objectiveevolutionaryalgorithm (SOEA) in designing and optimizing the morphology of a Six Articulated-Wheeled Robot (SAWR) is...
详细信息
ISBN:
(纸本)9781479957651
In this paper, the comparison of multi-objective evolutionary algorithm (MOEA) and Single-objectiveevolutionaryalgorithm (SOEA) in designing and optimizing the morphology of a Six Articulated-Wheeled Robot (SAWR) is presented. Results show that both methods are able to produce optimized SAWR which have smaller size with the capability to perform climbing motion. However, one of the solutions from the Pareto-set of MOEA is outperforming the fittest solution from SOEA. The solution is able to achieve the same performance of the fittest solution from SOEA and yet it is smaller in size. Besides that, another advantage of using MOEA is that MOEA is capable to produce a set of Pareto optimal solutions from the smallest SAWR with poor performance to the largest SAWR with robust performance which provide users a choice of solutions for trade-off between the two objectives.
The multi-objective evolutionary algorithm (MOEA) has shown remarkable capability of selecting feature subset. Most MOEAs use the cardinality of the feature subset as one of its objectives and adopt a strict Pareto do...
详细信息
The multi-objective evolutionary algorithm (MOEA) has shown remarkable capability of selecting feature subset. Most MOEAs use the cardinality of the feature subset as one of its objectives and adopt a strict Pareto dominance relationship to select individuals. However, these techniques limit available solutions and may omit several appropriate but dominated solutions. A multi-objective unsupervised feature selection algorithm (MOUFSA) is proposed to solve these issues. A new objective, which incorporates the correlation coefficient and cardinality of the feature subset, not only evaluates the redundancy of selected features but also provides several objective values for each particular size of feature subset. A relaxed archiving strategy based on negative epsilon-dominance and the box-based method is designed to preserve promising solutions even if they are dominated. Three new mutation operators of different abilities are also presented to enhance the algorithm. Nine UCI datasets and five fault recognition datasets are employed as test objects, and the obtained feature subsets are then used for subsequent classification and clustering. Experimental results show that MOUFSA outperforms several other multi-objective and traditional single-objective methods. (C) 2014 Elsevier B.V. All rights reserved.
Considering the trade-offs between conflicting objectives in project scheduling problems (PSPs) is a difficult task. We propose a new multi-objectivemulti-mode model for solving discrete time-cost-quality trade-off p...
详细信息
Considering the trade-offs between conflicting objectives in project scheduling problems (PSPs) is a difficult task. We propose a new multi-objectivemulti-mode model for solving discrete time-cost-quality trade-off problems (DTCQTPs) with preemption and generalized precedence relations. The proposed model has three unique features: (1) preemption of activities (with some restrictions as a minimum time before the first interruption, a maximum number of interruptions for each activity, and a maximum time between interruption and restarting);(2) simultaneous optimization of conflicting objectives (i.e., time, cost, and quality);and (3) generalized precedence relations between activities. These assumptions are often consistent with real-life projects. A customized, dynamic, and self-adaptive version of a multi-objective evolutionary algorithm is proposed to solve the scheduling problem. The proposed multi-objective evolutionary algorithm is compared with an efficient multi-objective mathematical programming technique known as the efficient epsilon-constraint method. The comparison is based on a number of performance metrics commonly used in multi-objective optimization. The results show the relative dominance of the proposed multi-objective evolutionary algorithm over the epsilon-constraint method. (C) 2013 Elsevier Ltd. All rights reserved.
This paper proposed a multi-objective evolutionary algorithm (MOEA) in designing the morphology of a six articulated-wheeled robot (SAWR) which has the ability to perform climbing motion. The first objective is to min...
详细信息
ISBN:
(纸本)9781479979103
This paper proposed a multi-objective evolutionary algorithm (MOEA) in designing the morphology of a six articulated-wheeled robot (SAWR) which has the ability to perform climbing motion. The first objective is to minimize the morphology design while the second objective is to maximize the performance of the SAWR in performing the climbing motion. Results show that the proposed MOEA is capable to produce a set of Pareto optimal solutions from the smallest SAWR with poor performance to the largest SAWR with robust performance. The Pareto set of optimal solutions provide users a choice of solutions for trade-off between the two objectives.
Uncertainties in design variables and problem parameters are often inevitable in multi-objective optimizations, and they must be considered in an optimization task if reliable Pareto optimal solutions are to be sought...
详细信息
Uncertainties in design variables and problem parameters are often inevitable in multi-objective optimizations, and they must be considered in an optimization task if reliable Pareto optimal solutions are to be sought. multi-objective reliability-based design optimization has been raised as a question in design for reliability, but the disadvantages of fixed evolutionary parameters, nonuniformly distributed Pareto optimal solutions and high computational cost hinder engineering applications of reliability-based design. To deal with it, this work proposes an integrated multi-objective cultural-based particle swarm algorithm to solve the double-loop reliability-based design optimization. In the inner optimization loop, the cultural space is composed of the elitism, situational and normative knowledge to adjust the parameters for swarm space, and the crowding distance ranking is introduced to update the global and local optimum and control the maximum number of solutions in elitism knowledge. The hybrid mean value method is improved to perform reliability analysis in the outer loop to suit both concave and convex types of performance functions. In addition, the car side-impact and the injection molding machine are chosen as multi-objective reliability design examples to demonstrate the effectiveness of the proposed approach. Simultaneously, results of car side-impact problem are compared with two traditional multi-objective reliability optimization algorithms, i.e., nondominated sorting genetic algorithm and crowding distance ranking-based multi-objective particle swarm optimizer, to assess the efficiency of the proposed approach. The results denote the proposed cultural-based multi-objective particle swarm optimizer is effective and feasible to solve the reliability-based design optimization problems.
暂无评论