Hybrid genetic algorithm and fuzzy clustering for bankruptcy prediction

In the design of a financial bankruptcy prediction model, financial ratio selection and classifier design play major roles. Methodology based on expert opinion, statistical theory and computational intelligence technique has been widely applied. In this study, a hybrid structure integrating statistical theory and computational intelligence technique was developed using genetic algorithm (GA) with statistical measurements and fuzzy logic based fitness functions for key ratio selection. A fuzzy clustering algorithm was used for the classifier design. In the experiments, two financial ratio sets, one extracted from the suggestions of other studies and the other obtained by using the GA toolbox in the SAS statistical software package, were applied to examine the proposed ratio selection schemes. For classifier design, the developed fuzzy classifier was compared with the well known BPNN classifier frequently used in other studies. Besides, comparison between the developed hybrid structure and other well applied structures was also given. Experimental results based on one to four years of financial data prior to the occurrence of bankruptcy were used to evaluate the performance of the proposed prediction model.     

Multi-objective optimization of stochastic disassembly line balancing with station paralleling

One of the major activities performed in product recovery is disassembly. Disassembly line is the most suitable setting to disassemble a product. Therefore, designing and balancing efficient disassembly systems are important to optimize the product recovery process. In this study, we deal with multi-objective optimization of a stochastic disassembly line balancing problem (DLBP) with station paralleling and propose a new genetic algorithm (GA) for solving this multi-objective optimization problem. The line balance and design costs objectives are simultaneously optimized by using an AND/OR Graph (AOG) of the product. The proposed GA is designed to generate Pareto-optimal solutions considering two different fitness evaluation approaches, repair algorithms and a diversification strategy. It is tested on 96 test problems that were generated using the benchmark problem generation scheme for problems defined on AOG as developed in literature. In addition, to validate the performance of the algorithm, a goal programming approach and a heuristic approach are presented and their results are compared with those obtained by using GA. Computational results show that GA can be considered as an effective and efficient solution algorithm for solving stochastic DLBP with station paralleling in terms of the solution quality and CPU time.     

A decision approach with multiple interactive qualitative objectives for product conceptual schemes based on noncooperative-cooperative game theory

Product development based on a morphological matrix involves the process of decision-based design. Although the decision process can generate conceptual schemes under the guidance of qualitative decision objectives, analysis of the interactions among the qualitative objectives is seldom considered, which can lead to unreliable optimal solutions by combining conflicting principle solutions. In addition, due to the ambiguity of the constraints among the qualitative objectives, multiple feasible schemes with equilibrium states are not considered in the concept decision stage. To solve these problems, a decision approach with multiple interactive qualitative objectives is developed for conceptual schemes based on noncooperative-cooperative game theory to consider the tradeoffs among objectives (e.g., cost, quality and operability) using discrete principle solution evaluation data. First, the morphological analysis method can obtain feasible schemes and determine the principle solutions for each subfunction. Second, the principle solutions are quantified using linguistic terms. Then, the subfunctions are categorized through cluster analysis to determine the suitable principle solution. Third, based on the clustering results, a noncooperative game decision model is constructed to identify multiple Nash equilibrium solutions that satisfy the constraints among the objectives. Fourth, a cooperative game decision model is constructed to obtain the optimal scheme as screened by the noncooperative game model. The case study proves that this approach can choose a relatively superior scheme under the existing technical conditions, thereby preventing inconsistency with the actual design expectations.     

MAGAD-BFS: A learning method for Beta fuzzy systems based on a multi-agent genetic algorithm

This paper proposes a learning method for Beta fuzzy systems (BFS) based on a multiagent genetic algorithm. This method, called Multi-Agent Genetic Algorithm for the Design of BFS has two advantages. First, thanks to genetic algorithms (GA) efficiency, it allows to design a suitable and precise model for BFS. Second, it improves the GA convergence by reducing rule complexity thanks to the distributed implementation by multi-agent approach. Dynamic agents interact to provide an optimal solution in order to obtain the best BFS reaching the balance interpretability-precision. The performance of the method is tested on a simulated example.     

Genetic algorithms for hardware–software partitioning and optimal resource allocation

A scheme for time and power efficient embedded system design, using hardware and software components, is presented. Our objective is to reduce the execution time and the power consumed by the system, leading to the simultaneous multi-objective minimization of time and power. The goal of suitably partitioning the system into hardware and software components is achieved using Genetic Algorithms (GA). Multiple tests were conducted to confirm the consistency of the results obtained and the versatile nature of the objective functions. An enhanced resource constrained scheduling algorithm is used to determine the system performance. To emulate the characteristics of practical systems, the influence of inter-processor communication is examined. The suitability of introducing a reconfigurable hardware resource over pre-configured hardware is explored for the same objectives. The distinct difference in the task to resource mapping with the variation in design objective is studied. Further, the procedure to allocate optimal number of resources based on the design objective is proposed. The implementation is constrained for power and time individually, with GA being used to arrive at the resource count to suit the objective. The results obtained are compared by varying the time and power constraints. The test environment is developed using randomly generated task graphs. Exhaustive sets of tests are performed on the set design objectives to validate the proposed solution.     

Solving a multi-floor layout design model of a dynamic cellular manufacturing system by an efficient genetic algorithm

This paper presents a mixed-integer programming model for a multi-floor layout design of cellular manufacturing systems (CMSs) in a dynamic environment. A novel aspect of this model is to concurrently determine the cell formation (CF) and group layout (GL) as the interrelated decisions involved in the design of a CMS in order to achieve an optimal (or near-optimal) design solution for a multi-floor factory in a multi-period planning horizon. Other design aspects are to design a multi-floor layout to form cells in different floors, a multi-rows layout of equal area facilities in each cell, flexible reconfigurations of cells during successive periods, distance-based material handling cost, and machine depot keeping idle machines. This model incorporates with an extensive coverage of important manufacturing features used in the design of CMSs. The objective is to minimize the total costs of intra-cell, inter-cell, and inter-floor material handling, purchasing machines, machine processing, machine overhead, and machine relocation. Two numerical examples are solved by the CPLEX software to verify the performance of the presented model and illustrate the model features. Since this model belongs to NP-hard class, an efficient genetic algorithm (GA) with a matrix-based chromosome structure is proposed to derive near-optimal solutions. To verify its computational efficiency in comparison to the CPLEX software, several test problems with different sizes and settings are implemented. The efficiency of the proposed GA in terms of the objective function value and computational time is proved by the obtained results.     

A breakable multi-item multi stage solid transportation problem under budget with Gaussian type-2 fuzzy parameters

This paper is based on two mathematical models for multi-item multi-stage solid transportation problem with budget on total transportation cost in Gaussian type-2 fuzzy environment considering the fixed opening charge and operating cost in distribution center. The first model is about transportation of breakable/damageable items, and the second one considers non breakable/damageable items. The main aspect here is to develop the mathematical formulation of multi stage related solid transportation problem where several items are available for transportation. In order to deal with the Gaussian type-2 fuzziness, two chance-constrained programming models are developed based on generalized credibility measures for the objective function as well as the constraints sets with the help of the CV-based reductions method. Finally the reduced model is turned into its equivalent parametric programming problem. The problem is of high complexity and is difficult to find the optimal solution by any classical method and hence a time and space based meta-heuristic Genetic Algorithm has been proposed. Also the equivalent crisp models are solved using GA and LINGO 13.0 and after comparison, GA results are better. The proposed models and techniques are finally illustrated by providing numerical examples. Some sensitivity analysis and particular cases are presented and discussed. Degrees of efficiency is also evaluated for both the techniques.     

Genetic algorithm-based identification of transfer function parameters for a rectangular flexible plate system

This paper focuses on an identification technique based on genetic algorithms (GAs) with application to rectangular flexible plate systems for active vibration control. A real coded GA with a new truncation-based selection strategy of individuals is developed, to allow fast convergence to the global optimum. A simulation environment characterizing the dynamic behavior of a flexible rectangular plate system is developed using the central finite difference (FD) techniques. The plate thus developed is excited by a uniformly distributed random disturbance and the input–output data of the system acquired is used for black-box modeling the system with the GA optimization using an autoregressive model structure. Model validity tests based on statistical measures and output prediction are carried out. The prediction capability of the model is further examined with unseen data. It is demonstrated that the GA gives faster convergence to an optimum solution and the model obtained characterizes the dynamic system behavior of the system well.     

使用遗传算法求解旅行商问题

遗传算法模拟自然界动物的繁衍来解决全局寻优问题,。旅行商问题是典型的组合优化问题,而且非常适于演示遗传算法的基本思想。对于遗传算法的优化,国内外许多专家结合旅行商问题提出了很多算子,结合其中的一些经典算子实现应用遗传算法解决旅行商问题的演示程序。     

Finite element modeling,characterization, and optimization design for the polymer-typed capacitive micro-arrayed ultrasonic transducer

This study constructs a two-dimensional axisymmetric finite element model (FEM) of the polymer-typed capacitive micromachined ultrasonic transducers (P-CMUT). The electromechanical FE model is created using the APDL programming technique. The ANSYS multiphysics solver with sequential approach and the Physics environment files are applied for the solution of the electrostatic-structural coupled-field analysis. Simulations are performed to investigate the operational characteristics, such as collapse voltage and the resonant frequency of the P-CMUT. The numerical results are found to be in good agreement with experimental observation. Having confirmed validity and accuracy of the proposed numerical model, the study of influence of each defined parameter on the collapse voltage and resonant frequency of the P-CMUT are also presented. To solve some conflict problems taken place in diverse physical fields, an integrated multi-objective design method involving electrical and mechanical characterization is developed to optimize the geometric parameters and material properties of the P-CMUT. The optimization search routine conducted using the genetic algorithm (GA) is connected with the commercial FEM software ANSYS to obtain the best design variable using multi-objective functions. The results show that the optimal parameter values satisfy the conflicting objectives of the design procedures, namely to minimize the collapse voltage while simultaneously maintaining a high resonant frequency. Overall, the presented results confirm that the combined FEM/GA optimization approach provides an efficient and versatile means for optimization design of the P-CMUT.     

Cellular genetic algorithm technique for the multicriterion design optimization

There have been increased activities in the study of genetic algorithms (GA) for problems of design optimization. The present paper describes a fine-grained model of parallel GA implementation that derives from a cellular-automata-like computation. The central idea behind the Cellular Genetic Algorithm approach is to treat the GA population as being distributed over a 2-D grid of cells, with each member of the population occupying a particular cell and defining the state of that cell. Evolution of the cell state is tantamount to updating the design information contained in a cell site, and as in cellular automata computations, takes place on the basis of local interaction with neighboring cells. A focus of the paper is in the adaptation of the cellular genetic algorithm approach in the solution of multicriteria design optimization problems. The proposed paper describes the implementation of this approach and examines its efficiency in the context of representative design optimization problems.     

Process industry scheduling optimization using genetic algorithm and mathematical programming

This article addresses the problem of scheduling in oil refineries. The problem consists of a multi-product plant scheduling, with two serial machine stages—a mixer and a set of tanks—which have resource constraints and operate on a continuous flow basis. Two models were developed: the first using mixed-integer linear programming (MILP) and the second using genetic algorithms (GA). Their main objective was to meet the whole forecast demand, observing the operating constraints of the refinery and minimizing the number of operational changes. A real-life data-set related to the production of fuel oil and asphalt in a large refinery was used. The MILP and GA models proved to be good solutions for both primary objectives, but the GA model resulted in a smaller number of operational changes. The reason for this is that GA incorporates a multi-criteria approach, which is capable of adaptively updating the weights of the objective throughout the evolutionary process.     

EVOLUTIONARY SEARCH FOR LIMIT CYCLE AND CONTROLLER DESIGN IN MULTIVARIABLE NONLINEAR SYSTEMS

A feature of many practical control systems is a Multi‐Input Multi‐Output (MIMO) interactive structure with one or more gross nonlinearities. A primary controller design task in such circumstances is to predict and ensure the avoidance of limit cycling conditions followed by achieving other design objectives. This paper outlines how such a system may be investigated using the Sinusoidal Input Describing Function (SIDF) philosophy quantifying magnitude, frequency and phase of any possible limit cycle operation. While Sinusoidal Input Describing function is a suitable linearization technique in the frequency domain for assessment of stability and limit cycle operation, it can not be employed in time domain. In order to be able to incorporate the time domain requirements in an overall controller design technique, the appropriate linearization technique suggested here is the Exponential Input Describing Function (EIDF). First, an evolutionary search based on a multi‐objective formulation is employed for the direct solution of the harmonic balance system matrix equation. The search is based on Multi‐Objective Genetic Algorithms (MOGA) and is capable of predicting specified modes of theoretically possible limit cycle operation. Second, the design requirements in time as well as frequency domain are formulated by a set of constraint inequalities. A numerical synthesis procedure also based on Multi‐Objective Genetic Algorithm is employed to adjust the initial compensator parameters to meet the imposed constraints. Robust stability and robust performance are investigated with respect to linearization uncertainty within the context of multiobjective formulation. In order to make the Genetic Algorithm (GA) search more amenable to design trade‐off between different and often contradictory specifications, a weighted sum of the functions is introduced. This criterion is subsequently optimized subject to the nonlinear system dynamics and a set of design requirements. Examples of use are given to illustrate the effectiveness of the proposed approach.     

Optimization of electronics component placement design on PCB using self organizing genetic algorithm (SOGA)

The optimal placement of electronic components on a printed circuit board (PCB) requires satisfying multiple conflicting design objectives as most of the components have different power dissipation, operating temperature, types of material and dimension. In addition, most electronic companies are currently emphasizing on designing a smaller package electronic system in order to increase the system performance. This paper presents a new self organizing genetic algorithm (SOGA) method for solving this multi-objective optimization problem. The SOGA can be viewed as a cascade of two GAs which consists of two steps fitness evaluation process to ensure that the fitness of selected chromosomes for each iteration process is optimally selected. The algorithm is developed based on weighted sum approach genetic algorithm (WSGA) where an inner loop GA is used to optimize the selection of weights of the WSGA. Experiments are conducted to evaluate the performance of SOGA. Four objective functions are formulated in the experiments which are temperature of components, area of PCB, high power component placement and high potential critical components distance. Comparisons of the performance of SOGA are made with two well known methods namely fixed weight GA (FWGA) and random weighted GA (RWGA). The results show that the SOGA gives a better optimal solution as compared to the other methods.     

A two-stage genetic algorithm for multi-objective job shop scheduling problems

This paper presents a two-stage genetic algorithm (2S-GA) for multi-objective Job Shop scheduling problems. The 2S-GA is proposed with three criteria: Minimize makespan, Minimize total weighted earliness, and Minimize total weighted tardiness. The proposed algorithm is composed of two Stages: Stage 1 applies parallel GA to find the best solution of each individual objective function with migration among populations. In Stage 2 the populations are combined. The evolution process of Stage 2 is based on Steady-State GA using the weighted aggregating objective function. The algorithm developed can be used with one or two objectives without modification. The genetic algorithm is designed and implemented with the GALIB object library. The random keys representation is applied to the problem. The schedules are constructed using a permutation with m-repetitions of job numbers. Performance of the proposed algorithm is tested on published benchmark instances and compared with results from other published approaches for both the single objective and multi-objective cases. The experimental results show that 2S-GA is effective and efficient to solve job shop scheduling problem in term of solution quality.     

A Novel Intelligent Multiobjective Simulated Annealing Algorithm for Designing Robust PID Controllers

This paper proposes an intelligent multiobjective simulated annealing algorithm (IMOSA) and its application to an optimal proportional integral derivative (PID) controller design problem. A well-designed PID-type controller should satisfy the following objectives: 1) disturbance attenuation; 2) robust stability; and 3) accurate setpoint tracking. The optimal PID controller design problem is a large-scale multiobjective optimization problem characterized by the following: 1) nonlinear multimodal search space; 2) large-scale search space; 3) three tight constraints; 4) multiple objectives; and 5) expensive objective function evaluations. In contrast to existing multiobjective algorithms of simulated annealing, the high performance in IMOSA arises mainly from a novel multiobjective generation mechanism using a Pareto-based scoring function without using heuristics. The multiobjective generation mechanism operates on a high-score nondominated solution using a systematic reasoning method based on an orthogonal experimental design, which exploits its neighborhood to economically generate a set of well-distributed nondominated solutions by considering individual and overall objectives. IMOSA is evaluated by using a practical design example of a super-maneuverable fighter aircraft system. An efficient existing multiobjective algorithm, the improved strength Pareto evolutionary algorithm, is also applied to the same example for comparison. Simulation results demonstrate high performance of the IMOSA-based method in designing robust PID controllers.     

A multi-objective evolutionary algorithm based QoS routing in wireless mesh networks

The huge demand for real time services in wireless mesh networks (WMN) creates many challenging issues for providing quality of service (QoS). Designing of QoS routing protocols, which optimize the multiple objectives is computationally intractable. This paper proposes a new model for routing in WMN by using Modified Non-dominated Sorting Genetic Algorithm-II (MNSGA-II). The objectives which are considered here are the minimization of expected transmission count and the transmission delay. In order to retain the diversity in the non-dominated solutions, dynamic crowding distance (DCD) procedure is implemented in NSGA-II. The simulation is carried out in Network Simulator 2 (NS-2) and comparison is made using the metrics, expected transmission count and transmission delay by varying node mobility and by increasing number of nodes. It is observed that MNSGA-II improves the throughput and minimizes the transmission delay for varying number of nodes and higher mobility scenarios. The simulation clearly shows that MNSGA-II algorithm is certainly more suitable for solving multiobjective routing problem. A decision-making procedure based on analytic hierarchy process (AHP) has been adopted to find the best compromise solution from the set of Pareto-solutions obtained through MNSGA-II. The performance of MNSGA-II is compared with reference point based NSGA-II (R-NSGA-II) in terms of spread.     

Decision support for conceptual bridge design

A knowledge based, decision support tool for the conceptual design of bridges is described. The system incorporates a restructured version of a knowledge base (KB) for conceptual bridge design. The restructuring results in a KB which can be readily altered and extended by system users who are not expert knowledge engineers. This is achieved by using a novel form of KB in which the knowledge is fragmented into separate concepts associated with design solutions. The system also includes a new style of user interface which provides a critiquing style of interaction with the KB only interacting with the user when it detects a possible error or a more suitable design solution. The new system has been developed using object oriented programming techniques which result in a structured and robust style of KB.     

A Hybrid Fuzzy Knowledge-Based Expert System and Genetic Algorithm for efficient selection and assignment of Material Handling Equipment

Material Handling (MH) is one of the key issues for every production site and has a great impact on manufacturing costs. The core concern in the design of a MH system is selecting the most suitable equipment for every MH operation and optimising them totally in order to attain an optimum solution. This paper presents a hybrid method for the selection and assignment of the most appropriate Material Handling Equipment (MHE). In the first phase, the system selects the most appropriate MHE types for every MH operation in a given application using a Fuzzy Knowledge-Based Expert System consisting of two sets of rules: Crisp Rules and Fuzzy Rules. In the second phase, a Genetic Algorithm (GA) searches throughout the feasible solution space, constituting of all possible combinations of the feasible equipment specified in the previous phase, in order to discover optimum solutions. The validity of the methodology developed in this paper is proved through the use of a real problem. Finally a comparison of the method with the other available publicised methods reveals the effectiveness of this hybrid approach.     

Sensor network optimization of gearbox based on dependence matrix and improved discrete shuffled frog leaping algorithm

This paper reports a new improved discrete shuffled frog leaping algorithm (ID-SFLA) and its application in multi-type sensor network optimization for the condition monitoring of a gearbox. A mathematical model is established to illustrate the sensor network optimization based on fault-sensor dependence matrix. The crossover and mutation operators of genetic algorithm (GA) are introduced into the update strategy of shuffled frog leaping algorithm (SFLA) and a new ID-SFLA is systematically developed. Numerical simulation results show that the ID-SFLA has an excellent global search ability and outstanding convergence performance. The ID-SFLA is applied to the sensor’s optimal selection for a gearbox. In comparison with GA and discrete shuffled frog leaping algorithm (D-SFLA), the proposed ID-SFLA not only poses an effective solving method with swarm intelligent algorithm, but also provides a new quick algorithm and thought for the solution of related integer NP-hard problem.