This paper describes a non-generational GA for multi-objective optimization problems (MOP) based on a crossover operator called DC (Dislocation Crossover). In it the replacement policy is such that an
This paper describes a non-generational GA for multi-objective optimization problems (MOP) based on a crossover operator called DC (Dislocation Crossover). In it the replacement policy is such that an
The New York city tunnel network problem is a classical case study for water distribution network rehabilitation or upgrading. However, standard problem formulations only offer single solutions based on fixed pressure...
详细信息
The New York city tunnel network problem is a classical case study for water distribution network rehabilitation or upgrading. However, standard problem formulations only offer single solutions based on fixed pressure head and demand requirements. This problem is re-examined through the use of multiple criteria optimization and a framework for interpreting these results is proposed. Results from previous studies were closely examined to highlight areas that can be improved. A 6-objective optimization problem was then formulated and results were compared with optimization using a single-objectivegeneticalgorithm (GA) and 2-objectives optimization using a non-dominated sorted geneticalgorithm-II. Although the initial runtime for the 6-objective optimization was long compared with single-objective GA, it provided solutions which span the whole spectrum of possibilities ( in terms of pressure head requirements and cost). It is concluded that the proposed 6-objectives formulation can indeed offer more design choices, hence flexibility, for designers and decision makers.
This paper describes an extension of a method for designing products with built-in disassembly means developed in our previous work, as applied to a realistic example of a desktop computer assembly. Given component ge...
详细信息
ISBN:
(纸本)1424400813
This paper describes an extension of a method for designing products with built-in disassembly means developed in our previous work, as applied to a realistic example of a desktop computer assembly. Given component geometries and revenues, the method simultaneously determines, through an optimization process, the spatial configuration of component, locator and fasteners such that the product can be most economically disassembled via a domino-like "self-disassembly" process triggered by the removal of one or a few fasteners. A multi-objective genetic algorithm is utilized to search for Pareto-optimal designs in terms Of four objectives: 1) satisfaction of the distance specification among components, 2) efficient use of locators on components, 3) profit of overall disassembly process, and 4) mass fraction of retrieved components. The method is applied to a simplified model of Power Mac G4 cube (R), and the results inspired a modification to the current design that can improve the ease of disassembly.
In this research, an approach is made to design machine cells using modular-reconfigurable machines to achieve certain characteristics of reconfigurable manufacturing. Each machine considered in the model consists of ...
详细信息
ISBN:
(纸本)9783901509469
In this research, an approach is made to design machine cells using modular-reconfigurable machines to achieve certain characteristics of reconfigurable manufacturing. Each machine considered in the model consists of some basic and auxiliary machine modules. By changing the auxiliary modules, different operations can be performed on the machines. A similarity measure among machines based on production flow information and auxiliary module requirement is developed. Machine cells are identified using a multi-objective evolutionary geneticalgorithm for a set of parts with parameters like volumes of production, alternative operation based process plans etc. The two objectives considered are minimization of inter-cell movement and total changes in auxiliary modules for the given production horizon. An illustrative problem and experimental results are given.
This paper presents a computational method for designing an assembly with multiple built-in disassembly pathways, each of which can be activated to retrieve certain components. It is motivated by the global sales of c...
详细信息
ISBN:
(纸本)0780394259
This paper presents a computational method for designing an assembly with multiple built-in disassembly pathways, each of which can be activated to retrieve certain components. It is motivated by the global sales of consumer products whose optimal end-of-life options vary geographically due to local recycling/reuse infrastructures and regulatory requirements. Given the sets of components to be retrieved at each location, the method simultaneously determines the spatial configurations of components and locator features, such that each set of desired components is retrieved via a domino-like "self-disassembly" process triggered by the removal of a fastener. A multi-objective generic algorithm is utilized to search for Pareto-optimal designs in terms of the realization of the desired disassembly pathways, the satisfaction of distance specifications among components, the minimization of disassembly cost at each location, and the efficient use of oncomponent locator features. A case study demonstrates the feasibility of the method.
Testing of VLSI circuits can cause generation of excessive heat which can damage the chips under test. In the random testing environment, high-performance CMOS circuits consume significant dynamic power during testing...
详细信息
ISBN:
(纸本)0780393635
Testing of VLSI circuits can cause generation of excessive heat which can damage the chips under test. In the random testing environment, high-performance CMOS circuits consume significant dynamic power during testing because of enhanced switching activity in the internal nodes. Our work focuses on the fact that power minimization is a Traveling Salesman Problem (TSP). We explore application of local search and geneticalgorithms to test set reordering and perform a quantitative comparison to previously used deterministic techniques. We also consider reduction of the original test set as a dual-objective optimization problem, where switching activity and fault coverage are the two objective functions.
＜正＞ Evolutionary design of large-scale structures has been the topic of much recent research;however, such designs are usually hampered by the time consuming stage of prototype evaluations using standard finite ele...
详细信息
＜正＞ Evolutionary design of large-scale structures has been the topic of much recent research;however, such designs are usually hampered by the time consuming stage of prototype evaluations using standard finite element analysis （FEA）. In this paper, a multi-objective metal-level （MOML） soft computing based evolutionary scheme is proposed. The neural network （NN） learns to approximate the fitness function in order to substitute the time consuming large-scale problem analysis （L-SPA） by FEA, the fuzzy controller updates parameters of a geneticalgorithm （GA） in order to balance exploitation vs. exploration in the search process, and the multi-objective GA optimizes parameters of membership functions in the fuzzy controller. The algorithm is first optimized on two benchmark problems, i.e. a 2-D Truss frame and an airplane whig. Then, robustness of the resulting optimization algorithm is tested on two other benchmark problems, i.e. a 3-layer composite beam and a piezoelectric bimorph beam. Performance of the proposed algorithm is compared with several other competing algorithms, i.e. a Fuzzy-GA-NN, a GA-NN and a simple GA in terms of both computational efficiency and accuracy. Statistical analysis indicates the superiority as well as robustness of the above approach as compared with the other optimization algorithms and applied to other benchmark problems. Specifically, the proposed approach finds better hardware designs more consistently while being computationally more efficient. Furthermore, these improvements are more pronounced as scale of the hardware design problem grows.
The multi-objective Job-Shop scheduling problem under multi-resource constrains was investigated and the active heuristic based operation precedence was developed to address the minimization of makespan, the minimizat...
详细信息
The multi-objective Job-Shop scheduling problem under multi-resource constrains was investigated and the active heuristic based operation precedence was developed to address the minimization of makespan, the minimization of total tardiness and the minimization of total idle time of the machines. Then a hybrid procedure was presented by combining the heuristic with multi-objective genetic algorithms and fuzzy evaluation. In the procedure, the representation based operation and the crossover operator based heuristics were adopted, which lead to chromosomes evolve in the active scheduling domain. The weights, which were specified randomly for each fuzzy evaluation, ensured the multidirectional search in the multi-objective genetic algorithm. The results of examples show that the procedure is available and efficient, and superior to the method of Ponnambala.
A multi-objective evolutionary search algorithm using a travelling salesman algorithm and geneticalgorithm for flow-shop scheduling is proposed in this paper. The initial sequence is obtained by solving the TSP. The ...
详细信息
A multi-objective evolutionary search algorithm using a travelling salesman algorithm and geneticalgorithm for flow-shop scheduling is proposed in this paper. The initial sequence is obtained by solving the TSP. The initial population of the geneticalgorithm is created with the help of a neighbourhood creation scheme known as a random insertion perturbation scheme, which uses the sequence obtained from TSP. The proposed algorithm uses a weighted sum of multiple objectives as a fitness function. The weights are randomly generated for each generation to enable a multi-directional search. The performance measures considered include minimising makespan, mean flow time and machine idle time. The performance of the proposed algorithm is demonstrated by applying it to benchmark problems available in the OR-Library.
暂无评论