Future products and processes will be impacted by biology and information technology. In this paper, the product and process design challenges are discussed. Evolutionary computation and data mining are two major tools that will cope with these challenges. The basic background of the two tools as well as examples demonstrating their applications in design and manufacturing are discussed.

This paper is a preliminary study on fault monitoring and on-line supervisor reconfiguration to achieve the fault-tolerant operation in cascaded manufacturing processes. To this end, we propose the on-line fault analysis scheme and the corresponding multiagent supervisory control policy by utilizing the index function defined at each state in the discrete event model of the unit processes. Based on the quantitative analysis of the abnormal status into the mutual fault and the essential failure, each agent supervisor of the unit process either makes the on- line fault-tolerant operation or the failure diagnosis in real- time.

Some inherent limitations exist in current Layered Manufacturing (LM) technologies (e.g. little choice of material, small part size and poor surface quality) and traditional NC machining (e.g. the confinement of tool accessibility to internal features). In order to overcome these limitations, a rapid manufacturing method called Robot based Layered Manufacturing (RoLM) in brief) is developed. A robot with a milling cutter mounted on the end- effector is used to build a part layer by layer. Given a part model, the determination of build orientation is the first step in the manufacturing cycle and has large effects on the surface quality and build time. In this paper, a set of criteria is proposed to rank the build orientations for RoLM by considering mainly part accuracy and build time. Algorithms are developed to calculate tool accessibility, part stability, and the number of required support for overhangs.

In order to succeed in today's global, competitive market, companies need continuous improvements in their product development processes. These improvements should result in expending fewer resources on the design process while achieving better quality. Automating the design process reduces resources needed and allows designers to spend more time on creative aspects that improve the quality of design. For the last three decades, engineers and designers have been searching for better ways to automate the product development process. For certain classes of design problems, which cover a large portion of real world design situations, the process can be automated using knowledge-based systems. These are design problems in which the knowledge sources are known in advance. Using techniques from Knowledge-Based Engineering, knowledge is codified and inserted into a knowledge-based system. The system activates the design knowledge, automatically generating designs that satisfy the design constraints. To increase the return on investment of building automated design systems, Knowledge management methodologies and techniques are required for capturing, formalizing, storing, and searching design knowledge.

Precision machining is becoming increasingly important in modern industry because many modern products require high form accuracy. An affordable approach to improve the accuracy of the surface profile of a workpiece is to adopt the on-line error forecasting and compensation control (FCC) techniques. In the present study, the consideration of variation of cutting force as a result of piezoactuator movement requires the formulation of ARMAX models. The time-series analysis based on ARMAX technique has an advantage over the traditional spectral method in that the latter can lead to the over-parameterization of the accompanying model. The roundness measurement results obtained from the practical experiments and the derived improvement percentages are grouped under one or more of the system parameters which include the ARMAX orders, feed rate, depth of cut, material, and forgetting factor. An expert system has been successfully developed to implement the rules using the Prolog language for helping the users to select suitable parameters for the FCC system of the lathe machine. Based on the measurement data, it can be shown that the lathe machine, when equipped with the ARMAX-based FCC system, can yield a minimum value of average improvement of 26% under the testing conditions.

The interest in consumer-focused virtual enterprises (VE) decision-making problem is growing fast. The purpose of this type of enterprise is to transform incomplete information about customer orders and available resources into-co-ordinated plans for production and replenishment of goods and services in the temporal network formed by collaborating units. This implies that information in the consumer-focused VE can be shared via Internet, Intranet, and Extranet for business-to-consumer (B2C), business-to-business service (B2B-S), and business-to-business goods (B2B-G) transactions. One of the goals of Internet-Based Management (e-management) is to facilitate transfer and sharing of data and knowledge in the context of enterprise collaboration. This paper discusses a generic framework of e-management that integrates intelligent information support group-decision making, and agreement modeling for a VE network. It offers the platform for design and modeling of diverse implementation strategies related to the type of agreement, optimization policies, decision-making strategies, organization structures, and information sharing strategies and mechanisms, and business policies for the VE.

The Stereolithography (STL) file format is a commonly-used format in the fields of Rapid Prototyping (RP) and Telemanufacturing and it is used to describe a 3D model that will be fabricated on an RP machine. Unfortunately, the format is known to be prone to error, and there is a need to investigate ways in which to check a file for anomalies and to correct any such anomalies. This is necessary to prevent the manufacturing of a flawed model thus saving the resources associated with a defective prototype. The errors that occur most commonly in the format can be divided into two groups, namely structural errors and geometrical errors. Structural errors occur on the file itself, where the file is either in the wrong format or does not comply with the STL standard. Geometrical errors are related to the model itself and include missing or duplicated triangles as well as incorrect surface normals due to calculation errors or incorrect triangle orientations. STLComplete, the software prototype developed by one of the authors, G.J. van Niekerk, in conjunction with his MSc dissertation, verifies the STL model for flaws and corrects less severe errors as well. Emphasis has been placed on the correction of faults and the presentation of errors to the user. The prototype will be given more attention here.

In recent times, while markets are reaching their saturation limits and customers are becoming more demanding, a paradigm shift has been taking place from mass production to mass- customized production (MCP). The concept of mass customization (MC) focuses on satisfying a customer's unique needs with the help of new technologies such as Internet, digital product realization, and re-configurable production facilities. In MC the needs of an individual customer are translated into design, accordingly produced, and delivered to the customer. In this research three hypothesis related to MCP are investigated by the data/information collected from ten companies, which are engaged in MCP. These three hypothesis are (1) mass-customized production systems can be classified into make-to-stock MCP, assemble-to-order MCP, make-to-order MCP, engineer-to-order MC, and develop-to-order MCP, (2) in mass-customized production systems the process of customization eliminates customer sacrifice, and (3) mass-customized production systems can deliver products at mass-production cost. The preliminary study indicates that while the first hypothesis is valid, MCP companies rarely fulfill what is stated in the other two hypotheses.

For make-to-order products in which the specification of some component-product variants are to be defined by customers, it is almost impossible to develop a bill-of-material system that supports customers' requests for quotation (RFQ) response activities using the existing bill-of-material concepts. To deal with this problem, a new technique for representing the relationship between parent-product variants and componenent- product variants in the generic bill-of-material (GBOM) structure is developed. The new technique is based on an assumption that the relationship between any two components in a product structure has pairwise independent property. The technique represents the relationship with compatibility constraints in the form of matrices and rules. The new representation provides users with a means to describing variants of make-to-order products as well as supporting the customers' RFQ response activities with only a limited amount of data which is comprehensive, easy to maintain, and easy to understand.

The definition of the agile paradigm has proved elusive and is often viewed as a panacea, in contention with more traditional approaches to operations strategy development and Larkin its own methodology and tools. The Theory of Constraints (TOC) is also poorly understood, as it is commonly solely associated with production planning and control systems and bottleneck management. This paper will demonstrate the synergy between these two approaches together with the Theory of Inventive Problem Solving (TRIZ), and establish how the systematic elimination of trade-offs can support the agile paradigm. Whereas agility is often seen as a trade-off free destination, both TOC and TRIZ may be considered to be route finders, as they comprise methodologies that focus on the identification and elimination of the trade-offs that constrain the purposeful improvement of a system, be it organizational or mechanical. This paper will also show how the TOC thinking process may be combined with the TRIZ knowledge based approach and used in breaking contradictions within agile logistics.

Parametric design methods are often used for the customization of geometric forms. In this paper, parametric design is approached from a different perspective. A fuzzy logic approach is developed to describe the geometry of a parametric model. Instead of using numerical numbers to describe a set of dimensions, the proposed method accepts imprecise linguistic input, such as large, small or medium large. The simplicity of such description facilitates and attracts even the unprofessional designers, or even household customers when such a system is available on the Internet. To demonstrate the proposed concept and method, a fuzzy parametric design program of wine glass is implemented. Both fuzzy description and numerical input are available and the preferred input is selectable by the user. When fuzzy input is selected, fuzzy reasoning is used to deduce the unimportant parameters so that a solid model of the wine glass can be generated in SolidWorks. Physical prototypes of a few wine glass designs are fabricated using rapid prototyping technologies.

Employment in the U.S. apparel industry has declined dramatically since the 1960s. Will it fall inexorably to zero, or is there some base level that can endure? If so, what strategic characteristics are required to survive? There is considerable interest in Quick Response Manufacturing (QRM), not only as a reason to support domestic manufacturing, but also as part of the larger goal of reducing supply chain costs. However, since Domestic Manufacturing is more expensive, why should anyone bother considering it? This paper presents an analytical model of a team approach that includes both domestic and offshore manufacturing. Despite the additional costs associated with U.S. manufacturing, our model predicts that including a domestic contractor is legitimate and cost effective. However, the alliance must be genuinely cooperative. A partnership has to be established early in the retailer's planning cycle, and the manufacturer should participate in the planning. Also, sharing data and making timely decisions imposes a strategic business approach, and the model allows us to describe the characteristic roles and capabilities required. Using this model for guidance, we anticipate that retailers will have the stock to satisfy more customers with fewer markdowns, while manufacturers will see increased margins and lower inventories.

This paper makes an investigation into combining heterogeneously- layered manufacturing (LM) machines into one homogeneous manageable unit, thereby allowing a user who wants to make use of LM to gain the advantages that LM provides. Such a user does not have to have direct access to LM hardware, but needs an Internet link to make use of the LM technology. The locating of production resources is done by a locating agent. This empowers small companies and inventors to get a conceptual design directly from the computer aided design file, thereby giving even the small designer or inventor the tools available to large companies.

SMEs play an important role in manufacturing industry. But from time to time there is a shortage in resources to complete the particular order in time. Number of systems is introduced to produce digital information in order to support product and process development activities. Main problem is lack of opportunity for direct data transition within design system modules when needed temporary extension of design capacity (virtuality) or to implement integrated concurrent product development principles. The planning experience in the field is weakly used as well. The concept of virtual manufacturing is a supporting idea to solve this problem. At the same time a number of practical problems should be solved like information conformity, data transfer, unified technological concepts acceptation etc. In the present paper the proposed ways to solve the practical problems of virtual manufacturing are described. General objective is to introduce the knowledge-based CAPP system as missing module for Virtual Manufacturing in the selected product domain. Surface-centered planning concept based on STEP- based modeling principles, and knowledge-based process planning methodology will be used to gain the objectives. As a result the planning module supplied by design data with direct access, and supporting advising environment is expected. Mould producing SME would be as test basis.

The objective of this research was to develop a decision support system (DSS) to study the impact of introducing new equipment into a medical apparel plant from a maintenance organizational structure perspective. This system will enable the company to determine if their capacity is sufficient to meet current maintenance challenges. The DSS contains two database sets that describe equipment and maintenance resource profiles. The equipment profile specifies data such as mean time to failures, mean time to repairs, and minimum mechanic skill level required to fix each machine group. Similarly, maintenance-resource profile reports information about the mechanic staff, such as number and type of certifications received, education level, and experience. The DSS will then use this information to minimize machine downtime by assigning the highest skilled mechanics to machines with higher complexity and product value. A modified version of the simplex method, the transportation problem, was used to perform the optimization. The DSS was built using the Visual Basic for Applications (VBA) language contained in the Microsoft Excel environment. A case study was developed from current existing data. The analysis consisted of forty-two machine groups and six mechanic categories with ten skill levels. Results showed that only 56% of the mechanic workforce was utilized. Thus, the company had available resources for meeting future maintenance requirements.

In recent years we have seen the exploration of the use of multiagent systems to provide control of manufacturing systems. This paper provides a brief review of agent-based systems applied to manufacturing control discussing the principle factors that contribute to defining the uniqueness of their design. The question arises in the development of such systems as to how much information about present processing and predicted future processing should be used in the negotiation and generation of plans for use in routing parts through the system. This paper explores this issue by first discussing the details of the negotiation mechanism employed. Then the results of tests are presented that compare the performance of the manufacturing system when the controlling agents are provided with the different forms of information.

Globalization can have a dramatic impact on manufacturing sector due to the fact that the majority of establishments in this industry are small to medium manufacturing companies. The role of Small and Medium Enterprises (SMEs) in the national economy has been emphasized all over the world, considering their contribution to the total manufacturing output and employment opportunities. The lack of marketing forces to regulate the operation of SMEs has been a fundamental cause of low efficiency for a long time. Computer Integrated Manufacturing (CIM) is emerging as one of the most promising opportunities for shrinking the time delays in information transfer and reducing manufacturing costs. CIM is the architecture for integrating the engineering, marketing and manufacturing functions through information system technologies. SMEs in general have not made full use of new technologies although their investments in CIM technology tended to be wider in scale and scope. Most of the SMEs only focus on the short-term benefit, but overlook a long- term and fundamental development on applications of new technologies. With the help of suitable information systems, modularity and low cost solutions, SMEs can compete in the global market. Considering the importance of marketing, information system, modularity and low cost solutions in the implementation of CIM in SMEs, a model has been developed and studied with the help of an empirical study conducted with British SMEs to facilitate the adoption of CIM. Finally, a summary of findings and recommendations are presented.

Vibration of a printed wiring board (PWB) based on Von Karman nonlinear strain field is analyzed. Equations of motion for the simply supported PWB and the clamped PWB are obtained for nonlinear vibration analysis. A 2-layer plastic PWB made of isotropic laminates is studied for its boundary condition effect on the vibratory behavior in deflection and stress distribution. Failure due to plane stress condition is estimated based on the composite failure criteria. Results are demonstrated through numerical computation. It is found that under the same loading, deflection of the clamped PWB is lower than that of the simply supported PWB, except at the lower frequency resonance. Nonlinear resonance occurs periodically with respect to the excitation frequency for both boundary conditions. At high excitation frequency, resonance deflection is mainly affected by the loading magnitude. Under the same load magnitude, simply supported PWB has the maximum stress close to that of the clamped PWB, and with improved reliability. Resonance in conjunction with stress analysis is critical in PWB failure prediction.

The high investment related to the acquisition of Flexible Manufacturing Systems forces firms to a better utilization of the machines. Different actions can be taken in order to avoid idle times of the machines: reduction of the unproductive times (time dedicated to rapid movements, tool exchange, pallet exchange, etc.), improvement of machines and, not last, a better management of the resources. The paper proposes a new policy for the management of tool operations in parallel machine FMS to minimize the idle times due to the lack of tools. The proposed policy uses new opportunities in manufacturing technology related with the use of network part programs in NC machines. It is already known in literature the potentiality of network part programs, more flexible than traditional sequential part programs that execute simply the rigid list of operations. Network part programs allow the different alternative ways to process each part. The way in which network part programs are executed by machines depends on the state of the tools and availability of the tools. The proposed method has been compared with other existing ones based on a real test case, a parallel machine FMS with two machines and a tool carrier.

A standard verbal description, which consists of function verbs, function adverbials and function sinks, is suggested for the common understanding of the design task. Three general purpose design catalogues: principle catalogue, module library and artifact store are introduced to arrange the design knowledge in accordance with the abstraction level of solutions. Based on the formal task description and the layered design catalogues, a systematic solution finding process is presented. The efficiency of a computer aided solution finding process can be drastically improved by using the constraints obtained from the function sinks. With the help of a component library the solutions, which are presented as a special function structure, can be adapted with the proven solutions. This adapted solution in form of connected available components can be further simulated and handed to the embodiment design. The whole process is explained with a design scenario of a gripper.

Parametric free forms are used in industries as disparate as footwear, toys, sporting goods, ceramics, digital content creation, and conceptual design. Optimizing tool path patterns and minimizing the total machining time is a primordial issue in numerically controlled (NC) machining of free form surfaces. We demonstrate in the present work that multi-cutter machining can achieve as much as 60% reduction in total machining time for compound sculptured surfaces. The given approach is based upon the pre-processing as opposed to the usual post-processing of surfaces for the detection and removal of interference followed by precise tracking of unmachined areas.

The application of Quad Flat Pack (QFP) and Ball Grid Array (BGA) technology in manufacturing of advanced electronic products has been grown significantly during the past several years. QFP and BGA has several advantages over traditional packaging technology in terms of smaller package area, higher I/Os, higher assembly yield, and so on. However, some potential problems of bridge among solder joints in QFP and creep-fatigue failure of solder joints in BGA serving occurs and often results in the failure of products. It has been found that both the bridge in QFP and fatigue failure of solder joints in BGA highly depends on the shape of solder joints. So, in this paper, the mathematical models of simulating the formation of solder joints in QFP and BGA are built based on minimum energy theorem, and the shape of solder joints are predicted. According to the predicted shape, the bridge mechanism in solder joints in QFP and the fatigue failure of solder joints in BGA is investigated.

Growing market turbulences and shorter product life cycles require a continuous adaptation of factory structures resulting in a continuous factory planning process. Therefore a new framework is developed which focuses on configuration and data management process integration. This enable an online system performance evaluation based on continuous availability of current data. The use of this framework is especially helpful and will guarantee high cost and time savings, when used in the early stages of the planning, called the concept or rough planning phase. The new framework is supported by a planning round table as a tool for team-based configuration processes integrating the knowledge of all persons involved in planning processes. A case study conducted at a German company shows the advantages which can be achieved by implementing the new framework and methods.

A new manufacturing methodology, termed shape-inclusive lay-up has been applied that allows the generation of three-dimensional preforms for the resin transfer molding (RTM) process. A flexible novel folding device for forming dry fabrics including non-crimp fabric (NCF) preform is designed and integrated with a Material Delivery System (MDS) into a robotic cell for manufacturing dry fiber composite aerospace components. The paper describes detailed design, implementation and operational performance of a prototype device. The proposed folding device has been implemented and tested by manufacturing a range of reinforcement structure preforms (C,T,J and I reinforcement preforms), normally used in aerostructure applications. A key advantage of the proposed device is its flexibility. The system is capable of manufacturing a wide range of components of various sizes without the need for reconfiguration.

The change in IT/IS strategy is about the Internet becoming a major part of the corporate environment and driving decisions more and more. Companies of all sizes and industries can fully engage employees, customers and partners to capitalize upon the new Internet economy. They can optimize supply chains, managing strategic relationships, reducing time to market, sharing vital information, and increasing productivity and shareholder value. Remaining competitive in today's rapidly changing global marketplace requires fast action. The problem is now how much, how soon, and what kind of Internet based components are essential for companies to be successful, and how the adoption of E-Integration can become a critical component of company's survival in an increasingly competitive environment. How information, knowledge and innovation processes can drive business success are fundamental notions for the information- based economy, which have been extensively researched and confirmed throughout the IT revolution. The new capabilities to use the Internet to supply large amounts of relevant information from multiple internal and external sources give the possibility to move from isolate Information Systems toward an integrated environment in every business organization. The article addresses how E-Integration must link together data from multiple sources, providing a seamless system, fully interoperable with pre-existing IT environment, totally scalable and upgradeable.

The concept behind improved enterprise resource planning systems (ERP) systems is the overall integration of the whole enterprise functionality into the management systems through financial links. Converting current software into real management decision tools requires crucial changes in the current approach to ERP systems. This evolution must be able to incorporate the technological achievements both properly and in time. The exploitation phase of plants needs an open web-based environment for collaborative business-engineering with on-line schedulers. Today's short lifecycles of products and processes require sharp and finely tuned management actions that must be guided by scheduling tools. Additionally, such actions must be able to keep track of money movements related to supply chain events. Thus, the necessary outputs require financial-production integration at the scheduling level as proposed in the new approach of enterprise management systems (ERM). Within this framework, the economical analysis of the due date policy and its optimization become essential to manage dynamically realistic and optimal delivery dates with price-time trade-off during the marketing activities. In this work we propose a scheduling tool with web-based interface conducted by autonomous agents when precise economic information relative to plant and business actions and their effects are provided. It aims to attain a better arrangement of the marketing and production events in order to face the bid/bargain process during e-commerce. Additionally, management systems require real time execution and an efficient transaction-oriented approach capable to dynamically adopt realistic and optimal actions to support marketing management. To this end the TicTacToe algorithm provides sequence optimization with acceptable tolerances in realistic time.

Assembly in the state when a product really comes to life and is able to function; a product achieves its functionality by the concerted efforts of its components. Assembling consists of the act of putting two or more components together; but behind the scenes work includes many activities necessary to realize a complete product, repair it, disassemble it for recycling or disposal and so on. During the past two decades or so, several methodologies- graph theory, matrix algebra, knowledge-based, object-oriented, etc. - have evolved for the modeling of assemblies. Each method has its own merit; but the suitability and practicality of a method often depends on the computer hardware capability and the state of available software language features (expression capability and libraries). Generally speaking, the modeling of assemblies involves a two step process: component relationship modeling and detailed CAD modeling. Product designers use the components relationship models, often informally, to complete their design. This paper describes the modeling techniques and implementation details pertaining to the component relations. Matrix algebra-based modeling framework that is used in this work is capable of representing product final assemblies, sub-assemblies and individual components. The framework has the required expression capability that may be used to perform bill of materials calculations, material requirements planning, the creation of assembly liaison or mating information, design for assembly (DFA) analysis and the generation of assembly plans/sequences.

In this paper, control chart pattern recognition using artificial neural networks is presented. An important motivation of this research is the growing interest in intelligent manufacturing systems, specifically in the area of Statistical Process Control (SPC). On-line automated process analysis is an important area of research since it allows the interfacing of process control with Computer Integrated Manufacturing (CIM) techniques. A back propagation artificial neural network is used to model X-bar quality control charts and identify process instability situations as specified by the Western Electric Statistical Quality Control handbook. Results indicate that the performance of the back propagation neural network is very accurate in identifying these control chart patterns. This work is significant in that the neural network output can serve as a link to process parameters in a closed-loop control system. In this way, adjustments to the process can be made on-line and quality problems averted.

The use of air bags in the presence of bad passenger and baby seat positions in car seats can injure or kill these individuals in case of an accident when this device is inflated. A proposed solution is the use of range sensors to detect passenger and baby seat risky positions. Such sensors allow the Airbag inflation to be controlled. This work is concerned with the application of different classification schemes to a real world problem and the optimization of a sensor as a function of the classification performance. The sensor is constructed using a new technology which is called Photo-Mixer-Device (PMD). A systematic analysis of the occupant detection problem was made using real and virtual environments. The challenge is to find the best sensor geometry and to adapt a classification scheme under the current technological constraints. Passenger head position detection is also a desirable issue. A couple of classifiers have been used into a simple configuration to reach this goal. Experiences and results are described.

This paper describes the work on the development of a group decision support system for customer driven product design. The customer driven is to develop products, which meet all customer requirements in whole life cycle of products. A process model of decision during product primary design is proposed to formulate the structured, semi-structured and unstructured decision problems. The framework for the decision support system is presented that integrated both advances in the group decision making and distributed artificial intelligent. The system consists of the product primary design tool kit and the collaborative platform with multi-agent structure. The collaborative platform of the system and the product primary design tool kit, including the VOC (Voice of Customer) tool, QFD (Quality Function Deployment) tool, the Conceptual design tool, Reliability analysis tool and the cost and profit forecasting tool, are indicated.

This work proposes the use of Timed Colored Petri nets as a formal base to a decision making tool for applications in industrial productive processes planning and programming. The Timed Colored Petri net is responsible for the transition of states in the decision process, establishing in time the use of resources and of heuristics that correspond to the more important managerial and operational actions for the planning activities and programming of the productive processes of an industrial plant. To negotiate with the uncertainties involved in a decision process, that in general takes care of the responsible specialist's knowledge for the routines involved in the productive system, we make use of the theory of fuzzy sets to suggest decisions logically consistent that obtain a viable solution just leading the viable states of the decision tree, that, in this case, is confused with the occurrence graph of the Petri net. As application example to the proposed model, we used a production system characterized by a port plant, whose model and simulation results are described at the end of this work.

The paper proposes the use of the Baysian Information Criterion (BIC), along with an algorithm to systematically select the appropriate structure of the backpropagation (BP) network for a given set of data. Simulation results with hydrological and economic data show that the algorithm performs very satisfactorily. Moreover, it compares with the method of Daqi and Shouyi for one hidden layer network and it is also used for the networks with more than one hidden layer.

The highest level of planning and control system is necessary, because production and logistics systems are not so flexible to follow, from day to day, sales evolutions. The companies are therefore held to standardize the good practices concerning the elaboration of their Sales, Inventories and Operations Planning (SIOP). The SIOP makes it possible to implement the strategic objectives defined by Top Management at the time of the Business Plan. It is the link between sales and manufacturing planning. The objectives of each of those depend on the specificity of their trade: the Sales Department will go for a maximum sales whereas Production will endeavor to keep industrial cost prices as low as possible while the Finance Department will try to optimize the use of available funds. There are several tools for this optimization: Graphical method and linear programming. Today, the economic context requires robust optimization.

A rule-based design software tool has been developed to specify the manufacturing process parameters to fabricate a contoured electron beam physical vapor deposited (EB-PVD) partially stabilized zirconia (PSZ) thermal barrier coating (TBC) on a curved surface. The expert data for this tool was developed from model predictions based on an experimentally verified, theoretical model of coating deposition rates as a function of position in the PVD vapor cloud (Knudsen cosine law). The predictive model was used for a wide variety of process parameters, including shadowing and non-uniform angular rotational velocity, to develop a database of contoured coating profiles (forward chaining). A coating profile matching routine was developed to identify the process parameters that yield a coating profile that matches the designers coating profile. The focus of this presentation will be on the software tool with emphasis on the expert database development and the coating thickness profile matching for a curved surface.

In this paper, we present a development of a framework for managing design and manufacturing process in a distributed environment. The framework offers the following facilities: (1) to represent the complicated engineering design processes (2) to coordinate design activities and execute the process in a distributed environment and (3) to support a collaborative design by sharing data and processes. In this paper, the process flow graphs, which consist in tasks and the corresponding input and output data, are used to depict the engineering design process on a process modeling browser. The engineering activities in the represented processes can be executed in a distributed environment through the cockpit of the framework. The communication among the related engineers to support a collaborative design is made on the collaborative design browser with SML underlying data structures of representing process information to make the framework extensible and platform- independent. The formal and flexible approach of the proposed framework to integrate the engineering design processes can be also effectively applied to coordinate concurrent engineering activities in a distributed environment.

In the field of nuclear plant installations a key problem is material flow control. To design the complex processes and security constraints of such a domain, we use an object oriented framework based on a role-task model, which is a generic organizational model, and descriptive classes relevant to the application domain. The object classes are created and instance libraries are elaborated allowing the user to build running scenarios composed of procedures, tasks and operations. The model is implemented in a case tool dedicated to dynamic modeling of the global installation. The system allows two manipulation levels: the expert level offers the resources to create the basic components and the related management primitives constituting a specific installation. These components are the metadata from which actual functional components are built. The user level implements the functions required to simulate various functional components are built. The user level implements the functions required to simulate various functional scenarios as well as possible diversions. The objective is to evaluate the time elapsed between the actual fraudulent action and its detection.

Perfect Distribution Conception, namely Ideal Ring Bundles can be used to find optimal solutions for manufacturing problems and advanced scientific researches in productivity enhancement based on combinatorial techniques. Ideal Ring Bundles (IRB) are cyclic sequences of integers which form perfect partitions of a finite interval (1,S) of integers. The sums of connected sub-sequences of IRB enumerate the set of integers (1,S) exactly R-times. Example : the IRB {1,7,2,3} containing four elements allows an enumeration of all numbers 1=1,2=2,3=3,4=3+1,5=2+3+1...13=1+7+2+3 exactly once (R=1). This property make IRBs useful in applications which need to partition sets with the smallest possible number of intersections. Application profiting from IRBs technique for example advanced manufacturing that has been shown in the paper.

The manufacture of upholstered furniture provides an excellent opportunity to analyze the effect of a comprehensive communication system on classical production management functions. The objective of the research is to study the scheduling heuristics that embrace the concepts inherent in MRP, JIT and TQM while recognizing the need for agility in a somewhat complex and demanding environment. An on-line, real-time data capture system provides the status and location of production lots, components, subassemblies for schedule control. Current inventory status of raw material and purchased items are required in order to develop and adhere to schedules. For the large variety of styles and fabrics customers may order, the communication system must provide timely, accurate and comprehensive information for intelligent decisions with respect to the product mix and production resources.

New technologies that are used on remediation sites rarely undergo testing for safety and health-related impacts on the workforce and the community. A user friendly safety and health assessment tool for design evaluation can help assure the safety for operators and the public. The system involves identifying a number of technology elements, the hazards associated with them, potential human injuries associated with the technologies, and the way they were designed. The system suggests recommendations for controlling the hazards and evaluates the interaction of technology elements. During the first year, the system was prototyped to a small existing technology of very limited scope. It showed that design for safety can indeed be possible using computer-systems. The focus of the second year has been to expand the system to accommodate 10 commonly used pieces of technologies. The system was developed to be a work-in-progress design aid.

In this paper a software tool for engineering design is presented, which supports the user in drafting automation tasks. The system was developed in cooperation with an industrial partner to assist sales representatives of components for automation (like linear drives, grippers, etc.). Today, sales representatives are supported with information on components in the form of paper catalogs or CD-ROMs, but although they do not sell complete systems, they need structured in formation on tasks and standards to meet their consulting obligation. Knowingplant provides a framework to structure and document automation knowledge in a hierarchical network of functions, where a function can be described as an automation task like temperature measuring. Each of these functions is either described by a set of sub function or can be refined by a decision between alternative functions. The decision making process is supported by decision tables and a search for similar decision problems on the base of parameter values. On the base of the hierarchical knowledge representation projects are drafted. Functions used in projects are automatically transferred together with their sub-functions until the system reaches a decision, which have to be made by the user. This leads to a fast structured draft of projects with all related subtasks and required parameters for further indoor processing. Knowingplant provides a flexible framework for system users as well as for knowledge authors who can easily update represented knowledge without programming, which enables a continuous growing knowledge base.

The potential application of case-based reasoning (CBR) in design support is illustrated through examples drawn from research at the University of Paisley, demonstrating the suitability of CBR for different aspects of design, different problem areas, and different design goals. A quality advisory system has been developed for the early stages of mechanical engineering design, the aim of which is to provide quality advice in a variant design situation. In the domain of software engineering CBR has been applied to advise on which metrics are appropriate fora assessing the quality of the software currently under design. The system integrates CBR with concepts from quality function deployment (QFD) and incorporates a case library holding past software quality histories. CBR has been applied in support of conceptual design: to capture detailed design histories by monitoring designer actions, and thereby support design reuse through the evaluation of designs, through the provision of query, browsing and replay facilities. The resulting system is aimed to support the design of safety critical systems, by assisting in the construction of safety arguments, and cooperative design.

Research on lifting activities has led to the design of several useful tools for evaluating tasks that involve lifting and material handling. The National Institute for Occupational Safety and Health (NIOSH) has developed a single task lifting equation. This formula has been frequently used as a guide in the field of ergonomics and material handling. While being much more complicated, the multi-task formula will provide a more realistic analysis for the evaluation of lifting and material handling jobs. A user friendly tool has been developed to assist professionals in the field of ergonomics in analyzing multitask types of material handling jobs. The program allows for up to 10 different tasks to be evaluated. The program requires a basic understanding of the NIOSH lifting guidelines and the six multipliers that are involved in the analysis of each single task. These multipliers are: Horizontal Distance Multiplier (HM), Vertical Distance Multiplier (VM), Vertical Displacement Multiplier (DM), Frequency of lifting Multiplier (FM), Coupling Multiplier (CM), and the Asymmetry Multiplier (AM). Once a given job is analyzed, a researched list of recommendations is provided to the user in an attempt to reduce the potential risk factors that are associated with each task.

Intelligent control of any manufacturing process requires an appropriate modeling of the process. Wire-cut EMD, being a highly complex process with non-linear behavior, is modeled by fuzzy logic approach in this paper . The knowledge base for the fuzzy model has been developed using the experimental results. Membership functions are defined for each parameter and a total of 10 input and 2 output parameters are considered to build the model. The results of the fuzzy model are compared with experimental values and a statistical analysis is presented.

This paper documents an Intelligent Manufacturing Workcell Controller (IMWC) environment for managing various critical aspects of a manufacturing can bottling production line. This paper focuses on the planning and control of a specific workcell to support flexible management. The system is described using an object-oriented modeling technique called Unified Modeling Language (UML). We further investigate the system's design architecture using a design pattern, specifically the mediator pattern.

Self-Assessment processes, initiated by a company itself and carried out by its own people, are considered to be the starting point for a regular strategic or operative planning process to ensure a continuous quality improvement. Their importance has increased by the growing relevance and acceptance of international quality awards such as the Malcolm Baldrige National Quality Award, the European Quality Award and the Deming Prize. Especially award winners use the instrument of a systematic and regular Self-Assessment and not only because they have to verify their quality and business results for at least three years. The Total Quality Model of the European Foundation for Quality Management (EFQM), used for the European Quality Award, is the basis for Self-Assessment in Europe. This paper presents a self-assessment supporting method based on a methodology of fuzzy control systems providing an effective means of converting the linguistic approximation into an automatic control strategy. In particular, the elements of the Quality Model mentioned above are interpreted as linguistic variables. The LR-type of a fuzzy interval is used for their representation. The input data has a qualitative character based on empirical investigation and expert knowledge and therefore the base- variables are ordinal scaled. The aggregation process takes place on the basis of a hierarchical structure. Finally, in order to render the use of the method more practical a software system on PC basis is developed and implemented.

Welding process parameters are indispensable to program arc welding robot. To simplify off-line programming (OLP) for robotic arc welding, we develop an arc welding expert system whcih can generate welding process parameters automatically. Its input data came from the feature database of welding part, which is set up by our feature modeling system. The expert system has become an important module of our RAWTOLPS (Robotic Arc Welding Task-level Off-Line System). It combines case-based reasoning with heuristic rule-based reasoning methods to deal with the welding process design. Moreover, artificial neural networks are introduced to the systems for reasoning and machine learning, and several network modules are developed to learn from welding process database, based on back-propagation neural networks. After some groups of actual welding process data were used to train the network models, several network models are established to both design the welding process and to predict the weld bead shape. Besides the ANN-based learning, cased-based learning are used in the expert system. These two methods have respectively their own characteristics, and can meet qualifications of different users. The experimental data show that the system can accomplish re-learning and expanding of welding process knowledge, and satisfy the command of the off-line programming system.

Based on the prime factors of multi-windows such as the shape, size, position, color, content and icons applied, the visualized user interface specification language (VUISL) is designed. The visualized user interface software generator (VUISG) is developed by YACC and C language. Thus the needed interface software can be generated automatically, whose universality and visualized ability are high.

Due to the development of advanced manufacturing technology and the introduction of Smart-Manufacturing notion in the field of modern industrial production, welding flexible manufacturing system (WFMS) using robot technology has become the inevitable developing direction on welding automation. In WFMS process, the flexibility for different welding products and the realizing on corresponding welding parameters control are the guarantees for welding quality. Based on a new intelligent arc-welding flexible manufacturing cell (WFMC), the system structure and control policies are studied in this paper. Aiming at the different information flows among every subsystem and central monitoring computer in this WFMC, Petri net theory is introduced into the process of welding manufacturing. With its help, a discrete control model of WFMC has been constructed, in which the system status is regarded as place and the control process is regarded as transition. Moreover, grounded on automation Petri net principle, the judging and utilizing of information obtained from welding sensors are imported into net structure, which extends the traditional Petri net concepts. The control model and policies researched in this paper have established foundation for further intelligent real-time control on WFMC and WFMS.

Information technology is driving improvements in manufacturing systems. Results are higher productivity and quality. However, corporate strategy is driven by a number of factors and includes data and pressure from multiple stakeholders, which includes employees, managers, executives, stockholders, boards, suppliers and customers. It is also driven by information about competitors and emerging technology. Much information is based on processing of data and the resulting biases of the processors. Thus, stakeholders can base inputs on faulty perceptions, which are not reality based. Prior to processing, data used may be inaccurate. Sources of data and information may include demographic reports, statistical analyses, intelligence reports (e.g., marketing data), technology and primary data collection. The reliability and validity of data as well as the management of sources and information is critical element to strategy formulation. The paper explores data collection, processing and analyses from secondary and primary sources, information generation and report presentation for strategy formulation and contrast this with data and information utilized to drive internal process such as manufacturing. The hypothesis is that internal process, such as manufacturing, are subordinate to corporate strategies. The impact of possible divergence in quality of decisions at the corporate level on IT driven, quality-manufacturing processes based on measurable outcomes is significant. Recommendations for IT improvements at the corporate strategy level are given.

The tailstock is used in turning operation to guide the workpiece rotation. This paper introduces an investigation for the influences of the tailstock axial force on the turning dynamics. A set of experiments were carried out to study the effect of different axial clamping force on the vibration of the tool- workpiece system and consequently on the workpiece roundness error. Vibrations generated during turning were measured in three directions (radial, axial, and tangential). Moreover, the resulted roundness errors on the workpieces surfaces were also measured. The results showed a great dependence of both vibrations existing during turning and the resulted roundness error, on the tailstock axial clamping force. The control of this force will improve the workpiece quality and detect the other errors such as chuck clamping. There is an optimum value for this force that reduces the roundness error to a minimum value. Also the vibration signals in the axial direction correlates comparatively well with the tailstock clamping force than those in other directions.

The design of fixtures for turbine blades is a difficult problem even for experience toolmakers. Turbine blades are characterized by complex 3D surfaces, high performance materials that are difficult to manufacture, close tolerance finish requirements, and high precision machining accuracy. Tool designers typically rely on modified designs based on experience, but have no analytical tools to guide or even evaluate their designs. This paper examines the application of kinematic algorithms to the design of six-point-nest, seventh-point-clamp datum transfer fixtures for turbine blade production. The kinematic algorithms, based on screw coordinate theory, are computationally intensive. When used in a blind search mode the time required to generate an actual design is unreasonable. In order to reduce the computation time, the kinematic methods are combined with genetic algorithms and a set of heuristic design rules to guide the search. The kinematic, genetic, and heuristic methods were integrated within a fixture design module as part of the Unigraphics CAD system used by Pratt and Whitney. The kinematic design module was used to generate a datum transfer fixture design for a standard production turbine blade. This design was then used to construct an actual fixture, and compared to the existing production fixture for the same part. The positional accuracy of both designs was compared using a coordinate measurement machine (CMM). Based on the CMM data, the observed variation of kinematic design was over two orders-of-magnitude less than for the production design resulting in greatly improved accuracy.