Variable neighborhood search for minimum cost berth allocation

The berth allocation problem is to allocate space along the quayside to incoming ships at a container terminal in order to minimize some objective function. We consider minimization of total costs for waiting and handling as well as earliness or tardiness of completion, for all ships. We assume ships can arrive at any given time, i.e., before or after the berths become available. The resulting problem, which subsumes several previous ones, is expressed as a linear mixed 0–1 program. As it turns out to be too time-consuming for exact solution of instances of realistic size, a Variable Neighborhood Search (VNS) heuristic is proposed, and compared with Multi-Start (MS), a Genetic Search algorithm (GA) and a Memetic Search algorithm (MA). VNS provides optimal solutions for all instances solved to optimality in a previous paper of the first two authors and outperforms MS, MA and GA on large instances.     

MATHEMATICAL ANALYSIS OF MUTATION OPERATOR AND ITS IMPROVED STRATEGY IN GENETIC ALGORITHMS

This paper analyzes the optimization problem of mutation probability in genetic algorithms by applying the definition of i-bit improved sub-space. Then fuzzy reasoning technique is adopted to determine the optimal mutation probability in different conditions. The superior convergence property of the new method is evaluated by applying it to two simulation examples.     

利用遗传算法由电阻抗计算四阶带通箱的低频响应

提出了一种基于电阻抗得到四阶带通扬声器系统低频频率响应的方法。该方法无需消声室,首先建立扬声器系统的电阻抗集中参数电路模型并测量得到电阻抗曲线,然后运用遗传算法优化模型中的元件参数值,使得由模型计算得到的阻抗曲线与测得的阻抗曲线相吻合,再根据模型计算得到低频响应曲线。测量结果表明理论曲线与实测曲线相吻合,说明基于电阻抗得到低频响应的方法是可行的和有效的。     

Product line selection and pricing analysis: Impact of genetic relaxations

A model for the product line selection and pricing problem (PLSP) is presented andthree solution procedures based on a genetic algorithm are developed to analyze the results based on consumer preference patterns. Since the PLSP model is nonlinear and integer, two of the solution procedures use genetic encoding to “relax” the NP hard model. The relaxations result in linear integer and shortest path models for the fitness evaluation which are solved using branch and bound and labeling algorithms, respectively. Performance of the quality of solutions generated by the procedures is evaluated for various problem sizes and customer preference structures. The results show that the genetic relaxations provide efficient and effective solution methodologies for the problem, when compared to the pure artificial intelligence technique of genetic search. The impact of the preference structure on the product line and the managerial implications of the solution characteristics generated by the genetic relaxations are also discussed. The models can be used to explicitly consider tradeoffs between marketing and operations concerns in designing a product line.     

Resolution approach for multi-objective problems with uncertain demands

The job-shop scheduling problem (JSP) is one of the hardest problems (NP-complete problem). In a lot of cases, the combination of goals and resource exponentially increases search space. The objective of resolution of such a problem is generally, to maximize the production with a lower cost and makespan. In this paper, we explain how to modify the objective function of genetic algorithms to treat the multi-objective problem and to generate a set of diversified “optimal” solutions in order to help decision maker. We are interested in one of the problems occurring in the production workshops where the list of demands is split into firm (certain) jobs and predicted jobs. One wishes to maximize the produced quantity, while minimizing as well as possible the makespan and the production costs. Genetic algorithms are used to find the scheduling solution of the firm jobs because they are well adapted to the treatment of the multi-objective optimization problems. The predicted jobs will be inserted in the real solutions (given by genetic algorithms). The solutions proposed by our approach are compared to the lower bound of the cost and makespan in order to prove the quality and robustness of our proposed approach.     

Fuzzy minimum weight edge covering problem

Minimum weight edge covering problem, known as a classic problem in graph theory, is employed in many scientific and engineering applications. In the applications, the weight may denote cost, time, or opponent’s payoff, which can be vague in practice. This paper considers the edge covering problem under fuzzy environment, and formulates three models which are expected minimum weight edge cover model, α-minimum weight edge cover model, and the most minimum weight edge cover model. As an extension for the models, we respectively introduce the crisp equivalent of each model in the case that the weights are independent trapezoidal fuzzy variables. Due to the complexity of the problem, a hybrid intelligent algorithm is employed to solve the models, which can deal with the problem with any type of fuzzy weights. At last, some numerical experiments are given to show the application of the models and the robustness of the algorithm.     

A new multi-objective genetic algorithm applied to hot-rolling process

A new genetic algorithms based multi-objective optimization algorithm (NMGA) has been developed during study. It works on a neighborhood concept in the functional space, utilizes the ideas on weak dominance and ranking and uses its own procedures for population sizing. The algorithm was successfully tested with some standard test functions, and when applied to a real-life data of the hot-rolling campaign of an integrated steel plant, it outperformed another recently developed multi-objective evolutionary algorithm.     

Crowding with nearest neighbors replacement for multiple species niching and building blocks preservation in binary multimodal functions optimization

This paper introduces a novel niching scheme called the q-nearest neighbors replacement (q-NNR) method in the framework of the steady-state GAs (SSGAs) for solving binary multimodal optimization problems. A detailed comparison of the main niching approaches are presented first. The niching paradigm and difference of the selection-recombination genetic algorithms (GAs) and the recombination-replacement SSGAs are discussed. Then the q-NNR is developed by adopting special replacement policies based on the SSGAs; a Boltzmann scheme for dynamically sizing the nearest neighbors set is designed to achieve a speed-up and control the proportion of individuals adapted to different niches. Finally, experiments are carried out on a set of test functions characterized by deception, epistasis, symmetry and multimodality. The results are satisfactory and illustrate the effectivity and efficiency of the proposed niching method.     

Simultaneous microemulsion liquid chromatographic analysis of fat-soluble vitamins in pharmaceutical formulations: Optimization using genetic algorithm

An environmentally benign and simple method has been proposed for separation and determination of fat-soluble vitamins using isocratic microemulsion liquid chromatography. Optimization of parameters affecting the separation selectivity and efficiency including surfactant concentration, percent of cosurfactant (1-butanol), and percent of organic oily solvent (diethyl ether), temperature and pH were performed simultaneously using genetic algorithm method. A new software package, MLR-GA, was developed for this purpose. The results indicated that 73.6 mM sodium dodecyl sulfate, 13.64% (v/v) 1-butanol, 0.48% (v/v) diethyl ether, column temperature of 32.5 °C and 0.02 M phosphate buffer of pH 6.99 are the best conditions for separation of fat-soluble vitamins. At the optimized conditions, the calibration plots for the vitamins were obtained and detection limits (1.06–3.69 μg mL⁻¹), accuracy (recoveries > 94.3), precision (RSD < 3.96) and linearity (0.01–10 mg mL⁻¹) were estimated. Finally, the amount of vitamins in multivitamin syrup and a sample of fish oil capsule were determined. The results showed a good agreement with those reported by manufactures.