Fractional cell formation in group technology using modified ART1 neural networks

Group technology (GT) is a manufacturing philosophy that attempts to reduce production cost by reducing the material handling and transportation cost. The GT cell formation by any known algorithm/heuristics results in much intercell movement known as exceptional elements. In such cases, fractional cell formation using reminder cells can be adopted successfully to minimize the number of exceptional elements. The fractional cell formation problem is solved using modified adaptive resonance theory1 network (ART1). The input to the modified ART1 is machine-part incidence matrix comprising of the binary digits 0 and 1. This method is applied to the known benchmarked problems found in the literature and it is found to be equal or superior to other algorithms in terms of minimizing the number of the exceptional elements. The relative merits of using this method with respect to other known algorithms/heuristics in terms of computational speed and consistency are presented.     

Automatic generation of group technology families during the part classification process

Initial part family formation and subsequent part classification are two important problems to be addressed in applying the group technology principle. Although these two problems are closely related, they have been treated separately. As an aggregate problem, the automatic creation of new part families during the classification process, is investigated. A two-layer neural network using the adaptive resonance theory is adopted. The capability of this neural network model of dealing with the stability-plasticity dilemma is utilised in classifying the parts into families and creating new families if necessary. A heuristic algorithm using the neural network is described, with illustrative examples.     

Manufacturing cell formation: A dual-objective simulated annealing approach

A dual-objective simulated annealing approach is developed for the purpose of concurrently forming machine cells and part families for cellular manufacturing. Simulated annealing has proved to be very effective in handling a variety of NP-hard problems. Originally, the simulated annealing algorithm was designed for applications with a single objective. For forming machine cells based on a machine-component incidence matrix, the result from single-objective simulated annealing was found to have room for improvement. In order to improve the performance of simulated annealing, a dual-objective approach is developed. The result clearly demonstrates the oustanding capabilities of the dual-objective approach. The discussions are illustrated with examples.     

Part family formation by memory association

Most of the part family formation methods are concerned with how to form the families as opposed to how to identify the families. However, a more appropriate approach would be to identify naturally occurring families since these methods are based on the production flow analysis, which uses already implemented routing data. This paper presents a new approach using the memory association of neural networks to identify naturally existing families. The developed system, Feature-Based Memory Association Network (FBMAN), operates by the exhaustive association approach which deals with the difficult problem of exceptional parts. Comparison with the results generated by other methods proves the effectiveness of FBMAN.     

Statistical process control with intelligence using fuzzy ART neural networks

With the automation development of manufacturing processes, artificial intelligence technology has been gradually employed to increase the automation and intelligence degree in quality control using statistical process control (SPC) method. In this paper, an SPC method based on a fuzzy adaptive resonance theory (ART) neural network is presented. The fuzzy ART neural network is applied to recognize the special disturbance of the manufacturing processes based on the classification on the histograms, which shows that the fuzzy ART neural network can adaptively learn the features of the histograms of the quality parameters in manufacturing processes. As a result, the special disturbance can be automatically detected when a feature of the special disturbance starts to appear in the histograms. At the same time, combined with spectrum analysis of the autoregressive model of quality parameters, the fuzzy ART neural network can also be utilized to adaptively detect the abnormal patterns in the control chart.     

Adaptive clustering algorithm for recycling cell formation: An Application of fuzzy ART neural networks

The recycling cell formation problem means that disposal products are classified into recycling part families using group technology in their end-of-life phase. Disposal products have the uncertainties of product status by usage influences during product use phase, and recycling cells are formed design, process and usage attributes. In order to deal with the uncertainties, fuzzy set theory and fuzzy logic-based neural network model are applied to recycling cell formation problem for disposal products. Fuzzy C-mean algorithm and a heuristic approach based on fuzzyART neural network is suggested. Especially, the modified FuzzyART neural network is shown that it has a good clustering results and gives an extension for systematically generating alternative solutions in the recycling cell formation problem. Disposal refrigerators are shown as examples.     

基于ART1神经网络的设计检索

深入研究了基于ART1神经网络的设计检索 ,并对ART1网络的算法进行了改进。改进后的算法可大大减少不必要的零运算 ,提高网络运行效率。与基于BP、Hopfield设计检索系统相比 ,ART1网络具有较大的优越性。主要体现在设计师可根据自己的意图 ,通过调节警戒参数得到所希望的检索结果。文中给出了齿轮类锻件的设计检索实例。     

Optimization and implementation of cellular manufacturing system in a pump industry using three cell formation algorithms

In recent years cellular manufacturing has become an effective tool for improving productivity. Attainment of full benefits of cellular manufacturing depends firstly on the design of the machine cells and part families and secondly on the method of operation which take full advantages of cell properties. Inappropriate methods of loading and scheduling can even lead to the failure of cellular manufacturing systems (CMS), however efficiently the cell is designed. This paper examines three array-based clustering algorithms, namely rank order clustering (ROC), rank order clustering-2 (ROC2) and direct clustering analysis (DCA) for manufacturing cell formation, with a real-life example to demonstrate the effectiveness of various clustering algorithms. The machine cell formation methods considered in this comparative and evaluative study belongs to the cluster formation approach of solving the MCF problem. The most effective method is selected and used to build the cellular manufacturing system. The comparison and evaluation are performed using four published performance measures and compares the improvements with the existing conventional system and the cellular manufacturing system. The above algorithms were written in the C++ language on an Intel/Pentium III-PC-compatible system.     

Adaptive control of nonlinear system using online error minimum neural networks

In this paper, a new learning algorithm named OEM-ELM (Online Error Minimized-ELM) is proposed based on ELM (Extreme Learning Machine) neural network algorithm and the spreading of its main structure. The core idea of this OEM-ELM algorithm is: online learning, evaluation of network performance, and increasing of the number of hidden nodes. It combines the advantages of OS-ELM and EM-ELM, which can improve the capability of identification and avoid the redundancy of networks. The adaptive control based on the proposed algorithm OEM-ELM is set up which has stronger adaptive capability to the change of environment. The adaptive control of chemical process Continuous Stirred Tank Reactor (CSTR) is also given for application. The simulation results show that the proposed algorithm with respect to the traditional ELM algorithm can avoid network redundancy and improve the control performance greatly.     

Adaptive nonlinear control using input normalized neural networks

An adaptive feedback linearization technique combined with the neural network is addressed to control uncertain nonlinear systems. The neural network-based adaptive control theory has been widely studied. However, the stability analysis of the closed-loop system with the neural network is rather complicated and difficult to understand, and sometimes unnecessary assumptions are involved. As a result, unnecessary assumptions for stability analysis are avoided by using the neural network with input normalization technique. The ultimate boundedness of the tracking error is simply proved by the Lyapunov stability theory. A new simple update law as an adaptive nonlinear control is derived by the simplification of the input normalized neural network assuming the variation of the uncertain term is sufficiently small.     

Distributed adaptive neural networks leader-following formation control for quadrotors with directed switching topologies

The leader-following formation problem is discussed for a team of quadrotors under directed switching topologies. To obtain a more general dynamic model, we describe the quadrotor system in a non-affine pure-feedback form with mismatched unknown nonlinearities. By employing an adaptive neural networks state observer to approximate the unknown nonlinear functions and to reconstruct the immeasurable inner states, we propose a novel distributed output feedback formation control protocol with the backstepping method combining with the dynamic surface control technique. From the Lyapunov stability theorem, all signals in the closed-loop formation system are proven to be cooperatively semiglobally uniformly ultimately bounded for any given bounded initial conditions. Finally, we proved that we verify the performance of the proposed formation control approach by a simulation study.     

Forming part-machine families for cellular manufacturing: A neural-network approach

This paper presents a new approach to forming group-technology part families for cellular manufacturing. The approach is based on neural-network technology, which mimics the way biological brain neurons perform to generate intelligent decisions. A procedure of forming part families using parallel and simple artificial neurons is described with examples. The implications and advantages of using neural networks in group technology are discussed.     

Nonlinear model identification and adaptive model predictive control using neural networks

This paper presents two new adaptive model predictive control algorithms, both consisting of an on-line process identification part and a predictive control part. Both parts are executed at each sampling instant. The predictive control part of the first algorithm is the Nonlinear Model Predictive Control strategy and the control part of the second algorithm is the Generalized Predictive Control strategy. In the identification parts of both algorithms the process model is approximated by a series-parallel neural network structure which is trained by a recursive least squares (ARLS) method. The two control algorithms have been applied to: 1) the temperature control of a fluidized bed furnace reactor (FBFR) of a pilot plant and 2) the auto-pilot control of an F-16 aircraft. The training and validation data of the neural network are obtained from the open-loop simulation of the FBFR and the nonlinear F-16 aircraft models. The identification and control simulation results show that the first algorithm outperforms the second one at the expense of extra computation time.     

A tradeoff analysis between process plan selection and cell formation in cellular manufacturing

In this paper, an algorithm to evaluate the tradeoff between conflicting objectives in process plan selection and cell formation is developed. Consideration of the minimisation of intercell material movement in cellular manufacturing is necessary but not in itself sufficient to produce a system for which the total work content is minimised. Solving the process plan selection and the cell formation problem for all possible alternative process plans is a time-consuming task, and therefore not economically justifiable. The algorithm is illustrated through the use of a sample problem that shows how it is possible to create a cell using the algorithm presented in this paper.     

基于移动机器人的多超声波传感器信息融合

智能化是未来机动机器人的发展方向,而智能化的基础是传感器技术的发展,机器人信息融合技术则弥补了使用单传感器带来的缺陷。本文利用多超声波传感器系统荻得移动机器人的环境信息,并使用人工神经网络对多传感器信息进行融合,从而达到对被测物体影像比较准确的认识,试验表明效果良好,具有很好的使用价值。     

Using neural networks for the diagnosis of localized defects in ball bearings

Two neural network based approaches, a multilayered feed forward neural network trained with supervised Error Back Propagation technique and an unsupervised Adaptive Resonance Theory-2 (ART2) based neural network were used for automatic detection/diagnosis of localized defects in ball bearings. Vibration acceleration signals were collected from a normal bearing and two different defective bearings under various load and speed conditions. The signals were processed to obtain various statistical parameters, which are good indicators of bearing condition, and these inputs were used to train the neural network and the output represented the ball bearing states. The trained neural networks were used for the recognition of ball bearing states. The results showed that the trained neural networks were able to distinguish a normal bearing from defective bearings with 100% reliability. Moreover, the networks were able to classify the ball bearings into different states with success rates better than those achieved with the best among the state-of-the-art techniques.     

A New Approach to Fuzzy Part-Family Formation in Cellular Manufacturing Systems

Group technology (GT) is one of the major issues in successful implementation of cellular manufacturing systems. The success of GT implementation depends only on an effective formation of part-families. The effective formation of part-families depends mainly on an effective formation of a similarity coefficient measure. Many similarity coefficients have been produced over the past three decades, but better similarity coefficient measures are required. The decision-making process in a manufacturing system often involves uncertainties and ambiguities. Under such circumstances, fuzzy methodologies have proved to be effective tools for taking fuzziness into consideration. In this paper, we use fuzzy cluster analysis to form part-families and assign parts to existing part-families. We have established a new approach to convert a fuzzy clustering matrix into a zero-one incidence matrix. We have also developed a new similarity coefficient measure and this coefficient measure is used to form a part-family. A mathemat-ical model that uses this similarity coefficient for solving optimally the part-family formation problems in cellular manufacturing is developed. Finally, it is compared with other models by giving an illustration with a numerical example.     

Quantification of the NA dependent change of shape in the image formation of a z-polarized fluorescent molecule using vectorial diffraction simulations

The point spread function of a fixed fluorophore with its dipole axis colinear to the optical axis appears donut-shaped when seen through a microscope, and its light distribution in the pupil plane is radially polarized. Yet other techniques, such as photolithography, report that this same light distribution in the pupil plane appears as a solid spot. How can this same distribution lead to a spot in one case but a donut in the other? Here, we show how the tube lens of the system plays a critical role in determining this shape. Using a vectorial treatment of image formation, we simulate the relative contributions of both longitudinal and radial components to the image of a dipole emitter and thus show how the donut (typically reported for z-polarized single molecule fluorescence microscopy) transforms into a solid spot (as commonly reported for photolithography) as the numerical aperture of the tube lens increases. We find that the transition point occurs around 0.7 NA, which is significantly higher than used for most microscopy systems and lower than for common photolithography systems, thus resolving the seeming paradox of dipole shape.     

The Role of Fuzzy and Genetic Algorithms in Part Family Formation and Sequence Optimisation for Flexible Manufacturing Systems

Group technology (GT) is one of the key issues in the successful implementation of flexible manufacturing systems (FMSs). The success of GT implementation is in the effective formation of part families (PFs) and similarity coefficients measures. Over the past three decades, many similarity coefficients have been proposed, but a better similarity coefficient measure is required. The decision-making process in a manufacturing system often involves uncertainties and ambiguities. Under such circumstances, fuzzy methodologies have proved to be an effective tool for taking fuzziness into consideration. The first part of this paper deals with the fuzzy part-family formation. This was achieved in the following ways: 1. A new similarity coefficient measure has been developed and this coefficient measure is used to form a part-family. 2. A mathematical model that uses this similarity coefficient for solving the part-family formation problems optimally in an FMS is developed. The fuzzy approach has the special advantage of producing more accurate results than conventional clustering and other methods. It not only reveals the specific part family that a part belongs to, but also provides the degree of membership of a part associated with each part family. This will give a balanced work load for the machine. In the second part of this paper, the introduction of the concept of genetic algorithms is proposed to eliminate more job sequences and, finally, the optimum sequence is obtained through the minimum penalty cost. Software is developed and implemented to obtain an optimum sequence and, finally, a numerical example is given as an illustration.