| 1 |
Front Matter |
|
|
Abstract
|
| 2 |
|
|
|
Abstract
|
| 3 |
Intelligent manufacturing systems |
Cihan H. Dagli |
|
Abstract
The quest for building systems that can function automatically has attracted a lot of attention over the centuries and created continuous research activity. As users of these systems we have never been satisfied; we demand more from the artifacts that are designed and manufactured. The current trend is to build autonomous systems that can adapt to changes in their environment. While there is much to be done before we reach this point it is not possible to separate manufacturing systems from this trend. The desire to achieve fully automated manufacturing systems is here to stay.
|
| 4 |
Intelligent systems architecture: Design techniques |
Deborah Stacey |
|
Abstract
The purpose in modern intelligent systems design is to specify, design and implement systems that have a high degree of machine intelligence. . can be defined as the ability to emulate or duplicate the sensory processing and decision making capabilities of human beings in computing machines (Barr and Feigenbaum, 1981). Intelligent systems need the ability to learn autonomously and to adapt in uncertain or partially-known environments if they are to progress past the academic domain and into a full engineering implementation. Different approaches have been utilized that either take advantage of one particular artificial intelligence methodology or exploit the complementary properties of several techniques to achieve a common goal.
|
| 5 |
Basic artificial neural network architectures |
Cihan H. Dagli,Pipatpong Poshyanonda |
|
Abstract
Different artificial neural network paradigms and architectures are proposed for the solution of manufacturing problems in the literature. A quick review of this literature highlights the fact that two paradigms (backpropagation and adaptive resonance theory (ART)) are used extensively to create intelligent manufacturing systems that can adapt to changing conditions. However, generally, these paradigms are not discussed in detail in papers if they are used in specific manufacturing application, so this chapter aims to fill this gap. After an introduction of basic artificial neural networks, three classical artificial neural network architectures — perceptron, backpropagation and adaptive resonance theory (ART1) — are examined.
|
| 6 |
Hybrid intelligent systems: Tools for decision making in intelligent manufacturing |
|
|
Abstract
|
| 7 |
|
|
|
Abstract
|
| 8 |
Conceptual design problem |
Ali Bahrami,Cihan H. Dagli |
|
Abstract
Design or problem solving is a natural human activity. We have been designing and acting as designers (sometimes unconsciously) throughout our lives. Design begins with the acknowledgment of needs and dissatisfaction with the current state of affairs and realization that some action must take place in order to correct the problem. When a small child moves a stool to an appropriate location so that she can use it to get to her toy, she has acted as a designer — of a rudimentary design — by positioning the stool so that she can satisfy her need of playing with the toy.
|
| 9 |
Machine-part family formation |
César O. Malavé,Satheesh Ramachandran |
|
Abstract
Group technology is a organizational approach which justifies small and medium batch sized production systems. Group technology based manufacturing systems offer the advantages of flow production as well as the production flexibility of batch manufacturing. By employing clustering techniques, the part-machine spectrum of the manufacturing system and the relevant manufacturing process are analyzed, and arranged according to design, similarity of machining and product flow. This leads to an organization of the production system into self-contained and self-regulating groups of machines called .. Each machine cell undertakes a maximal production of a family of parts having similar manufacturing characteristics. The principal advantages of a production system employing group technology lie in a substantially reduced work-in-process inventory and improved delivery performance because of reduced throughput time. The three general techniques of part family—machine group formation are:
|
| 10 |
Process planning |
Madhusudhan Posani,Cihan H. Dagli |
|
Abstract
Manufacturing planning, process planning, material processing, process engineering and machine routing are some of the names given to the topic referred to here as process planning.
|
| 11 |
Scheduling |
John Y. Cheung |
|
Abstract
An important aspect of automation in manufacturing systems is scheduling. The need for scheduling occurs at every level of the manufacturing process — from scheduling purchase of components and subcomponents, scheduling jobs and machines in the making and asembling process, to scheduling picking, packaging, shipping, etc. Furthermore, scheduling in material handling can occur at more than one level with varying degrees of detail and sophistication, e.g. a month-to-month schedule for orders and component purchase, a week-to-week schedule for components on the assembly line, a day-to-day schedule for jobs to be done in the machine shop, or an hour-by-hour schedule for each machine in the shop. With the emphasis of just-in-time manufacturing and production, scheduling is particularly important to ensure smooth operation of all phases of the entire manufacturing process.
|
| 12 |
Automated assembly systems |
Cihan H. Dagli,Mahesh Kumar Vellanki |
|
Abstract
Assembly is one of the major processes in a manufacturing environment, where many pre-fabricated parts are brought together within specified tolerances to produce a product which has some functional value. The need for sophisticated assembly systems that are intelligent and highly flexible has increased over the years and motivated many researchers to develop innovative techniques in implementing such systems. Today’s industry demands more flexibility in assembly automation due to rapid changes in product design imposed by highly competative global markets. Apart from flexibility, it is also desirable to build systems that can quickly adapt to new product designs and market changes. This fact was stated in an editorial by Dagli (1991) as:
|
| 13 |
Manufacturing feature identification |
Mark R. Henderson |
|
Abstract
Integration of CAD/CAM involves the use of a design description, design database model and applications while ensuring data consistency. The data types needed in a variety of applications may be different from the data type stored in a design database. In order to automate the CAD/CAM process, computer understanding of the meta-knowledge in the database or description and automatic data conversion to the next analysis stage are necessary (Woodwark, 1988).
|
| 14 |
Vision based inspection |
Joydeep Ghosh |
|
Abstract
In industrial manufacturing, product inspection is an important step in the production process. Since product reliability and quality management is of utmost importance in most mass-production facilities, 100% inspection of all parts, subassemblies, and finished products is often attempted. As a result, the inspection process is often the most costly stage in manufacturing.
|
| 15 |
Performance analysis of artificial neural network methods |
Benito R. Fernández Ph.D. |
|
Abstract
The complex demands of many modern applications cry out for flexible, adaptable devices that can quickly react to changing circumstances as solutions rather than mechanical brute force. Artificial neural systems (ANS) promise this capability. It is the intent of this chapter to realize ways to evaluate this accomplishment and its performance.
|
| 16 |
|
|
|
Abstract
|
| 17 |
Process monitoring and control |
Michel Guillot,Riadh Azouzi,Marie-Claude Cote |
|
Abstract
In everyday life, several mechanisms and processes surrounding us are monitored or controlled in view of achieving automated, efficient and more reliable operations. Process monitoring generally accounts for the . of events like process malfunctions, that require a specific and often discrete action. As opposed to monitoring, a controller . in order to regulate or control certain process outputs, or to optimize a given performance index. Thus process control usually involves the dynamics of the process, sensors and actuators in a closed-loop arrangement.
|
| 18 |
Adaptive control in manufacturing |
Yung C. Shin |
|
Abstract
Increasing requirements for high accuracy and productivity have forced many manufacturing firms to pursue automation. In order to meet the stringent requirements imposed on products, various forms of controllers have been used to regulate manufacturing processes and systems. Due to the extensive research on control theories, a wide variety of control schemes are now available. The controllers being used in practice, however, are mostly the types with fixed parameters designed by an engineer based on the physical system and specific control objective, which can address a single objective. This type of controller is acceptable for many simple mechanical systems. However, many manufacturing systems are inherently complex, non-linear, multivariable and subject to unknown external disturbance. When an autonomous operation of such a system is pursued, controllers must be able to adapt to the dynamic change of the system.
|
| 19 |
Fuzzy neural control |
L. H. Tsoukalas,A. Ikonomopoulos,R. E. Uhrig |
|
Abstract
Fuzzy neural control refers to the use of fuzzy logic and neural networks to control motors, actuators, and in general, the behavior of processes. The rapidly increasing number of fuzzy logic applications in process control and consumer electronics, as well as theoretical and hardware advances in neuro-control, provide significant incentives for studying fuzzy neural control. Of particular interest to control designers may prove the maturity of novel hardware systems, such as fuzzy logic and neural network boards, customized chips, and even fuzzy computers, where the data will be stored and arithmetically processed as fuzzy numbers.
|
| 20 |
Neural networks in continuous process diagnostics |
Najwa S. Merchawi,Soundar R. T. Kumara |
|
Abstract
In this chapter, the use of neural networks in diagnostics is investigated. Diagnostics is critical to any continuous process set up, to detect the occurrence of failures with accuracy and speed, particularly when failures can be catastrophic. Previous approaches to diagnostics problems have been rule-based. They suffered from slow response time and failure under missing information. Neural networks are fast because they use parallel processing concepts. They are tolerant of missing information because information is not stored locally but distributed over several nodes. In this research, a two-layer feedforward neural network is proposed as a modeling paradigm for continuous process diagnostics. Specific diagnostic problems in the domain of nuclear reactor control are studied. The diagnostic knowledge gathered by experts is represented in binary patterns. Different input and output representations are derived and the performance under each is analyzed. The results indicate that input representation is particularly important in determining the neural network’s recognition capability and the amount of tolerable noise in parameter readings.
|
发表于 2025-3-21 16:22:08
