This paper presents a framework for a collaboration platform that facilitates agile design process. The paper specifies the drivers for building such a collaboration platform, identifies its attributes, and proposes the mechanisms for resolving its dilemmas. The primary force that is driving agile product design is the market demand for the 'right products,' which have three attributes: (1) right features, (2) right time to market, and (3) right cost. The success of a company in marketplace is decided by how well it strikes a balance between these three attributes while developing new products. There have been several productivity boosting techniques such as CAD, CAM, CAE tools to assist designers at each stage of product development. However, the total product development process has not benefitted much from them, because of the inherent delays between the stages that account for 30 to 90 percent of the total product development time. An innovative approach, which employs web- based collaboration tools, can offer dramatic improvements in the process of introducing 'right products' into the market. The paper contends that an ideal collaboration platform should enable any designer located anywhere to design products using any CAD and any PDM on any platform. Such a collaboration platform potentially (1) reduces product development time, (2) curtails product development cost, and (3) improves the chances for first to market.

With the giant leaps that technology has taken in the past few years, some professions are still largely artistic. A good example of this is the profession of design. Designers produce products that are efficient, attractive, practical, but not safe. This is because most Universities around the world fail to include safety and health issues into their curriculum. To overcome that flaw, a rule-based tool was designed to assist designers in pointing out the unsafe conditions in their designs. This tool, that utilizes expert systems technologies, took five years to reach its current version. The tool is called TEXPERT. The latest version of TEXPERT underwent a number of changes and continued growth. This year's efforts resulted in the selection of a focus area and investigations into the selection of two technologies (demolition and decontamination- D&D) by which to validate the program. Validation and expansion of the rule-base continued with an emphasis on those components necessary for D&D. Research in object-oriented prototypes, determination of report-format and approach, and development of an initial visual project builder interface were also accomplished. Latest accomplishments included the development of new components, two-way interactions, and the implementation of a maintainable component database that the user interface can use to build the current library that will be available in the menu system.

The search for an increase in quality and speed of the diagnosis led to the reduction in the unavailability times of the production equipment. Thus, among the objectives of this study, need for operating on the level of the industrial sector of production through making of tools of assistance to the diagnosis of the abnormal operations, which are integrated into the environment of the system and allow the optimization of intervention times.

The genetic approach as the basis for the path search method for a multilink manipulator in complex environment has been discussed. To get over the major downfalls of the genetic approach the developed expert system (ES) to provide a specific purpose orientation at the stages of initial population formation and path crossing is proposed. We use the ES built of two modules: for forming recommendations for the initial population generation and for path crossing. The result for more than 4,000 experiments using five test workcells with various obstacles show that using the expert system increases the point attainability by 20 percent on the average.

The reduction of aluminum ore form alumina to metallic aluminum is done by dissolving the core in a cryolite bath at approximately 950 degrees C and passing DC current through the solution. Current flow through a typical cell can range from 75,000 to 200,000 Amperes, generating extremely strong electromagnetic fields. Common mode voltages of up to 800 VDC occur on reduction cell frames. Although the process is both complex and extremely energy intensive, it is generally only crudely instrumented because the hostile environment destroys any sensor placed in the bath. This paper describes a new approach to aluminum cell control employing the measurement of 'peripheral' process variables and the use of extensive modeling of the cell dynamics. A form of state observer is used to deduce critical process variables from the measurement of available related parameters. An experimental system is being installed at Century Aluminum and will be tested and refined over the coming year.

Bulk fertilizer storage companies have undergone dramatic changes including expansions and unification with various companies. This has resulted in the need for large, state- of-the-art fertilizer storage plants. Various factors such as proper layout of the facility, dry fertilizer storage, liquid fertilizer storage, bulk material handling equipment, automation of equipment automation etc. play a vital role in the development of storage plants. Planning for such a facility requires vast amounts of time and collaboration between the planning engineers, building contractor, equipment manufacturer, equipment automates, and other suppliers. Specializing exclusively in development of fertilizer storage plants, Stueve Construction Co. has long felt the need to develop a tool, which will aid in the proposal study and cost analysis of the various entities. This paper describes the feasibility study of developing an expert system incorporation the expertise of the various agents playing a pivotal role in the development of a bulk fertilizer storage plants. Apart from its use in developing a proposal study, its potential use as a marketing tool for the various agencies are also discussed.

This paper proposes a structured framework for exploiting data mining application for performance measures. The context is set in an airline company is illustrated for the use of such framework. The framework takes in consideration of how a knowledge worker interacts with performance information at the enterprise level to support them to make informed decision in managing the effectiveness of operations. A case study of applying data mining technology for performance data in an airline company is illustrated. The use of performance measures is specifically applied to assist in the aircraft delay management process. The increasingly dispersed and complex operations of airline operation put much strain on the part of knowledge worker in using search, acquiring and analyzing information to manage performance. One major problem faced with knowledge workers is the identification of root causes of performance deficiency. The large amount of factors involved in the analyze the root causes can be time consuming and the objective of applying data mining technology is to reduce the time and resources needed for such process. The increasing market competition for better performance management in various industries gives rises to need of the intelligent use of data. Because of this, the framework proposed here is very much generalizable to industries such as manufacturing. It could assist knowledge workers who are constantly looking for ways to improve operation effectiveness through new initiatives and the effort is required to be quickly done to gain competitive advantage in the marketplace.

This paper presents a modified Taguchi methodology to improve the robustness of modular product families against changes in customer requirements. The general research questions posed in this paper are: (1) How to effectively design a product family (PF) that is robust enough to accommodate future customer requirements. (2) How far into the future should designers look to design a robust product family? An example of a simplified vacuum product family is used to illustrate our methodology. In the example, customer requirements are selected as signal factors; future changes of customer requirements are selected as noise factors; an index called quality characteristic (QC) is set to evaluate the product vacuum family; and the module instance matrix (M) is selected as control factor. Initially a relation between the objective function (QC) and the control factor (M) is established, and then the feasible M space is systemically explored using a simplex method to determine the optimum M and the corresponding QC values. Next, various noise levels at different time points are introduced into the system. For each noise level, the optimal values of M and QC are computed and plotted on a QC-chart. The tunable time period of the control factor (the module matrix, M) is computed using the QC-chart. The tunable time period represents the maximum time for which a given control factor can be used to satisfy current and future customer needs. Finally, a robustness index is used to break up the tunable time period into suitable time periods that designers should consider while designing product families.

A decision support system using extended quality function deployment model (EQFDM) and internet application for manufacturing supply chain (SC) planning has been developed in this research. In this paper, a customer-focused quality evaluation approach, the EQFDM with internet application is employed to develop a coordinated planning system in SCs and assist mapping decisions of strategic planning into each partner's internal planning processes. To facilitate cooperation of SC partners in strategic planning, the hybrid planning process has been programmed into a web tool. The local planning has been supported by fuzzy logic approach so that approximate optimal solutions can be obtained avoiding difficulties of acquiring quantitative data. Through this intelligent Web based architecture, individual planning processes can be efficiently co-ordinated by means of efficient communication and visualizing consequences of a decision to be made on SC performance. Case study in a manufacturing (packaging) SC has been conducted to implement a scenario planning process for strategies on re-engineering the manufacturing SC. The research result shows that the intelligent system could be a promising tool for assisting strategic planning in a SC cooperation context.

This paper presents a new technique which uses a tree for constraining the compatibility among components in a logical bill of material (BOM) structure. The new representation for restricting possible combinations of components is designed to address the limitations of matrix representations that were used for the same purpose in earlier work. These matrix representations, which assume that the compatibility in a BOM can always be described for pairs of components, cannot be used for products in which the compatibility among three or more components is an issue. Thus, it is proposed to use a standard tree with a special representation, which is called a trie, to represent the compatibility between the components in a product configuration. Similar to the inter-component compatibility matrix, the trie can be used to validate and/or complete an arbitrary product configuration during the configuration process without having to spell out product configuration rules. The new representation provides the user with a means to describe all possible compatibility relationships with a limited amount of data. This trie representation is comprehensive, easy to maintain, and easy to understand.

The control of the aluminum reduction process is typically done based on measurements of cell voltage and current. Both signals are very noisy. Voltage fluctuations come from numerous sources such as CO₂ bubble generation, current fluctuations, metal pad motion, anode movement, electrical arcing, and changes in alumina concentration. The latter is an important control variable, but its effects on cell voltage are small and are often swamped by the signals from other sources. To date, control system designers have attempted to deal with this problem by extreme low-pass filtering and by suspension of all anode movements and alumina feed operations during the measurement period. This strategy is fairly effective, but leads in most case to sub- optimal cell operation most of the time. Research at West Virginia University and Century Aluminum seeks to improve upon the traditional strategies by incorporating knowledge about the process and prior cell control actions in the data analysis operation. For example, selective bandpass filtering and model-based feature detection strategies can be used to identify and isolate various portions of the composite signal. This paper discuses the methods being developed and the results of their application to this critical industrial process.

In polymer industries, the automation and control of reactors due to the progress in the areas of fuzzy control, neural networks, genetic algorithms, and expert systems lead to more secured and stable operation. When phenol and formaldehyde are mixed together, sudden heat is produced by the nonlinear exothermal reaction. Since sudden heat is liberated, polymerization process requires precise temperature control to avoid temperature run-away and the consequent damage to expensive materials. In practice, human involvement has been a source of errors that affects the quality of the product. This research proposes a design methodology for a sensor based computer control system. The duration of ON and OFF time of the relays is the parameters to be controlled in order to keep the exothermic reaction under control. This paper discusses a detailed simulation study of this exothermal process using MATLAB-SIMULINK-Fuzzy Logic toolbox. The model for the simulation study is derived from the constructed thermal system and responses are obtained. A predictive FLC structure is developed and compared to a classical PID control structure. Simulation results are obtained to ensure that the predictive FLC is better in controlling the reaction temperature.

Chemical Industries such as resin or soap manufacturing industries have reaction systems which work with at least two chemicals. Mixing of chemicals even at room temperature can create the process of exothermic reaction. This processes produces a sudden increase of heat energy within the mixture. The quantity of heat and the dynamics of heat generation are unknown, unpredictable and time varying. Proper control of heat has to be accomplished in order to achieve a high quality of product. Uncontrolled or poorly controlled heat causes another unusable product and the process may damage materials and systems and even human being may be harmed. Controlling of heat due to exothermic reaction cannot be achieved using conventional control methods such as PID control, identification and control etc. All of the conventional methods require at least approximate mathematical model of the exothermic process. Modeling an exothermal process is yet to be properly conceived. This paper discusses a design methodology for controlling such a process. A pilot plant of a reaction system has been constructed and utilized for designing and incorporating the proposed fuzzy logic based intelligent controller. Both the conventional and then an adaptive form of fuzzy logic control were used in testing the performance. The test results ensure the effectiveness of controllers in controlling exothermic heat.

Expert system development tool ESDT-HKD is a general-purpose language for knowledge engineering. Rule plus frame plus black board is used as knowledge structure of the system. It is a perfect system implemented on personal computer. There is particular knowledge structure, and generating and running environment for users. This article introduces the system architecture, knowledge structure and implementation.

In this paper, we analyze the situation of electronic commerce security and introduce the role of CA, then we describe the structure and function of CA. At last, we introduce the encryption algorithm. This CA can ensure security in the network transaction.

Modeling of welding process is the base of process control. Because welding process is a multivariable, strong coupling, time-varying and nonlinear system, traditional modeling methods are not suitable. In this paper, the dynamic neural network model for predicting backside width of pulsed GTAW weld pool by welding parameters and topside shape parameters was constructed. Orthogonal method was applied to design the sampling experiments. Experiments were carried on low carbon steel with 2mm thickness during pulsed gas tungsten arc butt-welding with gap. Based on self-developed vision sensor, double-side images of weld pool were captured simultaneously in a frame. By image processing the topside dimension and shape of weld pool, such as length, maximum width, gap width and the half-length ratio, and the backside dimension such as area, length and maximum width were calculated. Artificial neural network was applied to establish the model for predicting backside width of weld pool. The inputs of the model were the topside dimension, shape of weld pool and welding parameters such as pulse current, pulse duty ratio, and welding speed. The output of the model was the backside width of weld pool. The algorithm was the extended delta-bar-delta (EDD), and the learning ratio automatically determined by the algorithm. Threshold function was sigmoid function. The training cycle was selected to be 50000. The final EMS error of backside width was 5.2 percent. The simulation experiments were carried out to test the accuracy of the ANN model. From the results of the test, the output of ANN model can predict the backside width precisely.