Evolutionary Optimization Process Research Articles

Balancing exploration and exploitation with adaptive variation for evolutionary multi-objective optimization

Although recent studies have shown that evolutionary algorithms are effective tools for solving multi-objective optimization problems, their performances are often bottlenecked by the suitability of the evolutionary operators with respect to the optimization problem at hand and their corresponding parametric settings. To adapt the search dynamic of evolutionary operation in multi-objective optimization, this paper proposes an adaptive variation operator that exploits the chromosomal structure of binary representation and synergizes the function of crossover and mutation. The overall search ability is deterministically tuned online to maintain a balance between extensive exploration and local fine-tuning at different stages of the evolutionary search. Also, the coordination between the two variation operators is achieved by means of an adaptive control that ensures an efficient exchange of information between the different chromosomal sub-structures throughout the evolutionary search. Extensive comparative studies with several representative variation operators are performed on different benchmark problems and significant algorithmic performance improvements in terms of proximity, uniformity and diversity are obtained with the incorporation of the proposed adaptive variation operator into the evolutionary multi-objective optimization process.

An interactive evolutionary multi-objective optimization and decision making procedure

With the advent of efficient techniques for multi-objective evolutionary optimization (EMO), real-world search and optimization problems are being increasingly solved for multiple conflicting objectives. During the past decade of research and application, most emphasis has been spent on finding the complete Pareto-optimal set, although EMO researchers were always aware of the importance of procedures which would help choose one particular solution from the Pareto-optimal set for implementation. This is also one of the main issues on which the classical and EMO philosophies are divided on. In this paper, we address this long-standing issue and suggest an interactive EMO procedure which will involve a decision-maker in the evolutionary optimization process and help choose a single solution at the end. This study uses many year’s of research on EMO and would hopefully encourage both practitioners and researchers to pay more attention in viewing the multi-objective optimization as a aggregate task of optimization and decision-making.

Minimum-cost planning of the multimodal transport of pipes with evolutionary computation

Every day many kilometres of European highways are blocked by traffic jams. Congestion on roads and at air- ports adds the EU's fuel bill with a corresponding rise in pollution levels. In short, our present patterns of transport growth are unsustainable. One way of easing road congestion is to develop the efficient end-to-end movement of goods using two or more forms of transport in an integrated transport chain. We will focus on the multimodal transport problem that in- volves finding the most economical route in the distribution of cast iron ductile piping with or without mortar joint, both of different diameters and from different possible supply points to different points of destination over three transport net- works, road, rail and sea, which may have routes in common. The orders are made for quantities in linear metres of pipes. The economic cost of the transport on the various routes is dependent on the number of lorries, freight wagons and plat- forms required, and as these quantities must be obviously integer numbers. In practical applications the search space is di- mensionally very high, often there exist attractors into the search space due to the existence of multiple global optimum so- lutions, the cost function has discontinuities, many constraints, etc. The problem that has to be resolved is of great com- plexity, even using evolutionary algorithms. Our experience gained working several years in this optimization problem is described in this paper, to highlight that there is a need using evolutionary algorithms with certain learning ingredients and using strategies that progressively impose with more and more severity in the evolutionary optimization process the real scenario of the complex problem to get convergence to the global optimal solutions and simultaneously to obtain low com- putational cost total.

Multiobjective Optimization and Multiattribute Decision Making Study of Ship’s Principal Parameters in Conceptual Design

Numerous real-world problems relating to ship design are characterized by many alternatives as well as multiple conflicting objectives. Ship design is a complex endeavor requiring the successful coordination of many different disciplines, both technical and nontechnical. Conceptual design is the least defined stage of the ship design process and seeks to define the basic payloads and ship size characteristics. A hybrid approach for multiobjective optimization study of ship's principal parameters in conceptual design is proposed in the present analysis. In the first stage, a multiple objective genetic algorithm (MOGA) is employed to approximate the set of Pareto solution through an evolutionary optimization process. In the subsequent stage, a multiattribute decision making (MADM) approach is adopted to rank these solutions from best to worst and to determine the best solution in a deterministic environment with a single decision maker. A bulk carrier example, with 6 parameters, 3 criteria, and 14 constraints is conducted to illustrate the analysis process in present study. Pareto frontiers are obtained, and the ranking of the Pareto solution set is based on entropy weight and TOPSIS method. The ideal solution is compared with those from classic multiobjective methods.

A Competitive-Cooperative Coevolutionary Paradigm for Dynamic Multiobjective Optimization

In addition to the need for satisfying several competing objectives, many real-world applications are also dynamic and require the optimization algorithm to track the changing optimum over time. This paper proposes a new coevolutionary paradigm that hybridizes competitive and cooperative mechanisms observed in nature to solve multiobjective optimization problems and to track the Pareto front in a dynamic environment. The main idea of competitive-cooperative coevolution is to allow the decomposition process of the optimization problem to adapt and emerge rather than being hand designed and fixed at the start of the evolutionary optimization process. In particular, each species subpopulation will compete to represent a particular subcomponent of the multiobjective problem, while the eventual winners will cooperate to evolve for better solutions. Through such an iterative process of competition and cooperation, the various subcomponents are optimized by different species subpopulations based on the optimization requirements of that particular time instant, enabling the coevolutionary algorithm to handle both the static and dynamic multiobjective problems. The effectiveness of the competitive-cooperation coevolutionary algorithm (COEA) in static environments is validated against various multiobjective evolutionary algorithms upon different benchmark problems characterized by various difficulties in local optimality, discontinuity, nonconvexity, and high-dimensionality. In addition, extensive studies are also conducted to examine the capability of dynamic COEA (dCOEA) in tracking the Pareto front as it changes with time in dynamic environments.

Simulation Optimization in Uncertain Environments: An Evolutionary Approach

The design of manufacturing systems often involves simulation optimization approaches to search for the best system performance. Metaheuristic approaches are more and more used to optimize simulation models. In most existing approaches, the optimization is performed with fixed environmental conditions (e. g., part demand, breakdown rates are assumed to be perfectly known). However, in practice the actual data about the system may differ from those used in the simulation model (e. g., modification of the part mix). To cope with this, the candidate solutions, in a simulation optimization process, can be compared on various possible environments, using such principles as those proposed by Taguchi. Unfortunately, when metaheuristics are used, simulation optimization can be very much time consuming, since each solution has to be compared on a number of different environments. In order to provide robust solutions in a more reasonable time, we propose a two stage heuristic search. First, a restricted set of n promising solutions is identified using an evolutionary multimodal simulation optimization process, using the concept of base environmental scenario, recently published. Then robustness evaluation on many environments is performed only on these n promising solutions and the most robust can be chosen. This approach is illustrated on a supply chain problem where several parameters have to be defined. As a result, 13 solutions are found. The most robust solution is not the one that yields the best results in the environment assumed by the decision maker. This result shows how important it is to be able to consider several environments in the simulation optimization process.

RETRACTED: Study of multi-objective optimization and multi-attribute decision-making for economic and environmental power dispatch

Environmental awareness and the recent environmental policies have forced many electric utilities to restructure their practices to account for their emission impacts. One way to accomplish this is by reformulating the traditional economic dispatch problem such that emission effects are included in the mathematical model. The economic/environmental dispatch problem is a multi-objective non-linear optimization problem with constraints. This study presents a hybrid approach to solve the combined economic-emission dispatch problem (CEED). In the first stage, a non-dominated sortinggenetic algorithm II (NSGA II) is employed to approximate the set of Pareto solution through an evolutionary optimization process. In the subsequent stage, a multi-attribute decision-making (MADM) approach is adopted to rank these solutions from best to worst and to determinate the best solution in a deterministic environment with a single decision maker. This hybrid approach is tested on a six-unit system to illustrate the analysis process in present analysis. Pareto frontiers are obtained and the ranking of Pareto solutions is based on entropy weight and TOPSIS method. Results obtained show that the hybrid approach has a great potential in handling multi-objective optimization problem.

Network Topologies Emerging in An Evolutionary Optimization Process

Network Topologies Emerging in An Evolutionary Optimization Process

Evolution and Learning Mediated by Differences in Developmental Timing

The present paper introduces a mutation-based evolutionary algorithm that evolves genes to regulate the developmental timings of phenotypic values. For each generation, an individual in the evolutionary algorithm time-sequentially generates a given number of entire phenotypes before finishing its life. Each gene represents a cycle time of changing probability for determining its corresponding phenotypic value, which is an indicator of developmental timing. In addition, the algorithm has a learning mechanism such that, during the lifetime of an individual, genes representing a long cycle time can change the probability of adaptation more easily than genes representing a short cycle time. Therefore, if the diversity of the genes is maintained, it can be expected that the algorithm provides a different evolution speed to each phenotypic value. The present paper also discusses a new approach to depicting an evolutionary optimization process. An evolutionary optimization process involves the identification of linkage between variables, and therefore, network structures formed using the identified linkage information determine how the evolutionary algorithm solves a given optimization problem. The proposed approach regards an evolutionary optimization process as a change in the network topology that emerges in the process of linkage identification. The simulation results indicate that evolution and learning mediated by the difference in developmental timing helps to sequentially solve hard uniformly-scaled bit optimization problems with linkage between variables.

A hybrid approach for multi-objective combinatorial optimisation problems in ship design and shipping

Numerous real-world problems relating to ship design and shipping are characterised by combinatorially explosive alternatives as well as multiple conflicting objectives and are denoted as multi-objective combinatorial optimisation (MOCO) problems. The main problem is that the solution space is very large and therefore the set of feasible solutions cannot be enumerated one by one. Current approaches to solve these problems are multi-objective metaheuristics techniques, which fall in two categories: population-based search and trajectory-based search. This paper gives an overall view for the MOCO problems in ship design and shipping where considerable emphasis is put on evolutionary computation and the evaluation of trade-off solutions. A two-stage hybrid approach is proposed for solving a particular MOCO problem in ship design, subdivision arrangement of a ROPAX vessel. In the first stage, a multi-objective genetic algorithm method is employed to approximate the set of pareto-optimal solutions through an evolutionary optimisation process. In the subsequent stage, a higher-level decision-making approach is adopted to rank these solutions from best to worst and to determine the best solution in a deterministic environment with a single decision maker.

Design and statistical properties of robust functional networks: A model study of biological signal transduction

A simple flow network model of biological signal transduction is investigated. Networks with prescribed signal processing functions, robust against random node or link removals, are designed through an evolutionary optimization process. Statistical properties of large ensembles of such networks, including their characteristic motif distributions, are determined. Our analysis suggests that robustness against link removals plays the principal role in the architecture of real signal transduction networks and developmental genetic transcription networks.

Strong effects of network architecture in the entrainment of coupled oscillator systems

Random networks of coupled phase oscillators, representing an approximation for systems of coupled limit-cycle oscillators, are considered. Entrainment of such networks by periodic external forcing applied to a subset of their elements is numerically and analytically investigated. For a large class of interaction functions, we find that the entrainment window with a tongue shape becomes exponentially narrow for networks with higher hierarchical organization. However, the entrainment is significantly facilitated if the networks are directionally biased--i.e., closer to the feedforward networks. Furthermore, we show that the networks with high entrainment ability can be constructed by evolutionary optimization processes. The neural network structure of the master clock of the circadian rhythm in mammals is discussed from the viewpoint of our results.

Optimisation multi-objectif d'un instrument robotisé à haute mobilité pour la chirurgie mini-invasive

Robotics plays a central role in minimalty invasive surgery. However existing robotic systems lack extra intra-cavital motion capabilities. They can be improved by the use of of kinematically redundant instruments whose design can't result from a simple intuitive process. The research carried out refers to the optimal design of high-mobility surgical instruments dedicated to minimally-invasive surgery. The proposed methodology is based on a multi-objective evolutionary optimisation process, coupled with a realistic simulation of the surgical task. This design methodology has been applied to a coronary artery bypass grafting procedure. It results in a robotized surgical instrument made of 9 degrees offreedom: DRIMIS (Dextrous Robotic Instrument for Minimally Invasive Surgery).

Modelling external bone adaptation using evolutionary structural optimisation

External remodelling is significant in the bone healing process, and it is essential to predict the bone external shape in the design of artificial bone grafts. This paper demonstrates the effectiveness of the evolutionary structural optimisation (ESO) method for the simulation of bone morphology. A two-dimensional ESO strategy is developed which is capable of finding the modified bone topology beginning with any geometry under any loading conditions. The morphology of bone structure is described by the quantitative bone adaptation theory, which is integrated with the finite element method. The evolutionary topology optimisation process is introduced to find the bone shape. A rectangle, which occupies a larger space than the external shape of the bone structure, is specified as a design domain; the evolutionary process iteratively eliminates and redistributes material throughout the domain to obtain an optimum arrangement of bone materials. The technique has been tested on a wide range of examples. In this paper, the formation of trabecular bone architecture around an implant is studied; as another example, the growth of the coronal section of a vertebral body is predicted. The examples support the assertion that the external shape of bone structure can be successfully predicted by the proposed ESO procedure.

Evolutionary structural optimization of steel gusset plates

Evolutionary structural optimization is applied to determine the optimum shape of steel gusset plates subjected to axial forces. A number of different gusset plates used in various types of connections is considered for this purpose. The evolutionary structural optimization approach is employed to find the optimum shapes of a gusset plate used in these connections. The first example considers a gusset plate having two holes which are utilized in the connection of double angle carrying a tensile force. Within this example the effect of certain parameters in the evolutionary structural optimization process such as material removal ratio, mesh size and modeling of holes on the final shape is investigated. The gusset plates having three, four and five holes are also designed for optimal shape. Furthermore design examples include two rows of multiple holes as well as staggered holes and connections with multiple members. The final shapes obtained in the single member bolted connections are generally similar to those used in practice. However, they are lighter. Those shapes obtained for the multiple member connections and welded double angle connections are unpredictable. Although the shapes obtained in all the examples are lighter than the ones used in practice, they might be more expensive to produce. It is shown that the evolutionary structural optimization method has a potential in determining the optimum shape of gusset plates.

Determination of strut-and-tie models using evolutionary structural optimization

This paper introduces a method to determine strut-and-tie models in reinforced concrete (RC) structures using evolutionary structural optimization (ESO). Even though strut-and-tie models are broadly adapted in the design of reinforced concrete members subjected to shear and torsion, conventional methods can hardly offer useful models of RC members subjected to complex loadings and geometry conditions. In this paper, the basic idea of the ESO method is used to determine more rational strut-and-tie models. Since an optimum topology of structures, finally obtained by the ESO method, usually represents a truss-like structure, the ESO method using truss elements can effectively be used in finding the best strut-and-tie models in RC structures. To prevent the structural instability that may occur during the evolutionary optimization process, a brick element composed of six truss elements is designed as a basic element unit. Systematic removal of each brick element that has the least virtual strain energy follows, and the optimal load transfer mechanism of an RC structure, which is equivalent to the optimal topology of the strut-and-tie model, is finally characterized on the basis of an optimization criterion of minimizing the total elastic strain energy of the structure. Several RC structures are used as examples to demonstrate the capability of the proposed method in finding the best strut-and-tie model of each RC structure and to verify its efficiency in application to real design problems.

Computational Evolutionary Optimization of Red Phosphor for Use in Tricolor White LEDs

An evolutionary optimization process involving a genetic algorithm and combinatorial chemistry was employed in an attempt to develop titanate-based red phosphors suitable for use in tricolor white light-emitting diodes. An eight-cation oxide system, including (K,Li,Na)x(Y,Gd,La,Eu)yTizOδ, was screened in terms of luminescence efficiency. The use of a combination of a genetic algorithm and combinatorial chemistry enhanced the efficiency of searching, as it is applied to the screening of phosphors in a limited composition space. As a result, the optimum composition was determined to be (Na0.92Li0.08)(Y0.8Gd0.2)TiO4:Eu3+, the reproducibility of which was achieved by the conventional solid-state reaction method. Using the (Na0.92Li0.08)(Y0.8Gd0.2)TiO4:Eu3+ phosphor and an InGaN chip, a red LED was realized, so that the applicability of this new phosphor to white LEDs was confirmed. The luminescence of this phosphor was 110% that of the well-known scheelite variant phosphor and 600% that of a commercially avai...

Parisian camera placement for vision metrology

This paper presents a novel camera network design methodology based on the Parisian evolutionary computation approach. This methodology proposes to partition the original problem into a set of homogeneous elements, whose individual contribution to the problem solution can be evaluated separately. A population comprised of these homogeneous elements is evolved with the goal of creating a single solution by a process of aggregation. The goal of the Parisian evolutionary process is to locally build better individuals that jointly form better global solutions. The implementation of the proposed approach requires addressing aspects such as problem decomposition and representation, local and global fitness integration, as well as diversity preservation mechanisms. The benefit of applying the Parisian approach to our camera placement problem is a substantial reduction in computational effort expended in the evolutionary optimization process. Moreover, experimental results coincide with previous state of the art photogrammetric network design methodologies, while incurring in only a fraction of the computational cost.

Genetic Algorithm−Assisted Combinatorial Search for a New Green Phosphor for Use in Tricolor White LEDs

An evolutionary optimization process involving a genetic algorithm and combinatorial chemistry was employed for the development of green phosphors which are suitable for tricolor white-light-emitting diodes. To accomplish a high luminescent efficiency at 400 nm excitation, we screened a seven-cation oxide system including Tb, Gd, Ce, Mg, Si, Al, and B. The combination of a genetic algorithm and combinatorial chemistry enhanced the searching efficiency when applied for phosphor screening. As a result, the optimized composition was Tb(0.01)Gd(0.02)Ce(0.04)B(0.1)Si(0.83)O(delta), The luminance of this borosilicate glass was 67% that of ZnS:Cu,Al at 400-nm excitation.

Evolutionary design and behavior analysis of neuromodulatory neural networks for mobile robots control

Evolutionary Robotics (ER) is one of promising approaches to design robot controllers which essentially have complicated and/or complex properties. In most ER research, the sensory-motor mappings of robots are represented as artificial neural networks, and their connection weights (and sometimes the structure of the networks) can be optimized in the parameter spaces by using evolutionary computation. However, generally, the evolved neural controllers could be fragile in unexperienced environments, especially in real worlds, because the evolutionary optimization processes would be executed in idealized simulators. This is known as the gap problem between the simulated and real worlds. To overcome this, the author focused on evolving an on-line learning ability instead of weight parameters in a simulated environment. According to recent biological findings, actually, the kinds of on-line adaptation abilities can be found in real nervous systems of insects and crustaceans, and it is also known that a variety of neuromodulators (NMs) play crucial roles to regulate the network characteristics (i.e. activating/blocking/changing of synaptic connections). Based on this, a neuromodulatory neural network model was proposed and it was utilized as a mobile robot controller. In the paper, the detail behavior analysis of the evolved neuromodulatory neural network is also discussed.