Simplification Methodology Research Articles

A Methodology for simplification and interpretation of backpropagation-based neural network models

A new methodology for building inductive expert systems known as neural networks has emerged as one of the most promising applications of artificial intelligence in the 1990s. The primary advantages of a neural network approach for modeling expert decision processes are: (1) the ability of the network to learn from examples of experts' decisions that avoids the costly, time consuming, and error prone task of trying to directly extract knowledge of a problem domain from an expert and (2) the ability of the network to handle noisy, incomplete, and distorted data that are typically found in decision making under conditions of uncertainty. Unfortunately, a major limitation of neural network-based models has been the opacity of the inference process. Unlike conventional expert system decision support tools, decision makers are generally unable to understand the basis of neural network decisions. This problem often makes such systems undesirable for decision support applications. A new methodology is presented that allows the development of highly simplified backpropagation neural network models. This methodology simplifies netw variables that are not contributing to the networks ability to produce accurate predictions. Elimination of unnecessary input variables directly reduces the number of network parameters that must be estimated and consequently the complexity of the network structure. A primary benefit of this development methodology is that it is based on a variable importance measure that addresses the problem of producing an interpretation of a neural network's functioning. Decision makers may easily understand the resulting networks in terms of the proportional contribution each input variable is making in the production of accurate predictions. Furthermore, in actual application the accuracy of these simplified models should be comparable to or better than the more complex models developed with the standard approach. This new methodology is demonstrated by two classification problems based on sets of actual data.

Concurrent engineering: Automation, tools, and techniques

From the Publisher: Concurrent Engineering: Automation, Tools, and Techniques illustrates how to design products, components or systems, across a range of industries, while taking into account such constraints as the ability to manufacture, reliability requirements, quality, cost and customer needs. In long-awaited, much-needed contrast to other books, which focus on a single aspect, such as managing the design process or product reliability, Concurrent Engineering covers all the aspects of the field. This comprehensive analysis includes in-depth treatment of how the consequences of various technical solutions are assessed in all life-cycle phases of a product; automatic idealization control for analysis performed during concurrent engineering design; an intelligent interactive CAD system which has a mechanism for real-time constraint checking in the design process; conceptual design with physical features and an aspect model integration using a metamodel; automated assembly planning, an important integrative issue in concurrent engineering; the three stages of the design process: system design, parameter design and tolerance design and the methodology for simplification of the design process. In the final chapter, the preliminary design of a light utility helicopter is presented as a case study of how concurrent design can be accomplished by applying ten general characteristics of a concurrent engineering process. Figures, photographs, charts and tables are generously provided throughout to clarify the text and an extensive bibliography is provided for extended study.

Application of a power plant simplification methodology: the example of the condensate feedwater system

A novel framework for the systematic simplification of power plant design is described with a focus on the application for the optimization of condensate feedwater system (CFWS) design. The evolution of design complexity of CFWS is reviewed with emphasis upon the underlying optimization process. A new evaluation methodology which includes explicit accounting of human as well as mechanical effects upon system availability is described. The unifying figure of merit for an operating system is taken to be net electricity production cost. The evaluation methodology is applied to the comparative analysis of three designs. In the illustrative examples, the results illustrate how inclusion in the evaluation of explicit availability related costs leads to optimal configurations. These are different from those of current system design practices in that thermodynamic efficiency and capital cost optimization are not overemphasized. Rather a more complete set of design-dependent variables is taken into account, and other important variables which remain neglected in current practices are identified. A critique of the new optimization approach and a discussion of future work areas including improved human performance modeling and different optimization constraints are provided.

Model of the use of coal, gas, and oil

A methodology for model simplification is described, which uses the pseudo-data approach and known statistical techniques. An analysis of pseudo-data generated from the Brookhaven Energy System Optimization Model (BESOM) was undertaken to determine the interrelationship of imported oil use with coal and gas use in that model. The simplified model is tested by projecting the imported oil needed for levels of coal and gas use consistent with the national restrictions embodied in BESOM. Projections of the derived functions were verified with full-scale BESOM runs. The effects of gas and coal prices on oil use are shown.