An expert system approach for designing air-conditioning systems for cars and trucks is presented. A brief introduction to the automotive application of the vapor-compression refrigeration cycle is provided as general background. The method presented uses an integrated approach combining the power of conventional analysis programs, databases and model-based expert system technology. Some sample rules from the knowledge base have been included in the paper to illustrate the application of the domain knowledge and its interaction with algorithmic programs. The system architecture is very open and modular, and it lends itself to easy modifications and future expansions. Possibilities for system enhancements are also outlined in the paper. The approach presented in this paper provides substantial benefits to the automotive air-conditioning design engineer particularly in the early stages of new vehicle platform planning and development. A pilot system has been successfully tested at General Motors for preliminary design of automotive air-conditioning systems.

Studies on the application of AI techniques to engineering design, analysis, and manufacturing (AI EDAM) problems have been expanding steadily over the past decade. These studies have led to the founding of many new journals and the initiation of series of conferences. If all these research efforts were successful, some of us might have been unemployed but living happily off the royalties from our successful research products that were deployed in practice.

The linking of research in machine learning with research in knowledge-based design is such that each of the two areas benefit from the consideration of the other. The use of machine learning in design addresses the perceived need to support the capture and representation of design knowledge, because handcrafting a representation is a difficult and time-consuming task. In addition, design provides a task with which to investigate the usefulness of existing machine learning techniques, and, perhaps, to discover new ones.

This study focuses on the application of neural network techniques to adaptive remeshing of an idealized squareshaped structure and individual triangle by using triangular elements. The backpropagation learning algorithm is implemented by a supervised training technique to deal with the problem of remeshing structural elements in a structural analysis. A recent study introduced a structural remeshing that incorporated a finite-element analysis with an adaptive mesh generation technique. The main objective of the study is to demonstrate how neural networks can be employed to remesh structural elements without using numerically intensive computations. One essential requirement of this approach is the selection of feasible and appropriate training and testing data. The exploration of neural network in remeshing structural elements is a fundamental technique that looks beyond the finite-element and adaptive mesh generation techniques. It also demonstrates the capability of neural network to represent n-dimensional space and track each individual characteristic in that separate space. In general, an overview is presented, and the potential of neural networks to use the backpropagation algorithm instead of the more conventional approach of numerical methods is demonstrated.

This paper focuses on that form of learning that relates to exploration, rather than generalization. It uses the notion of exploration as the modification of state spaces within which search and decision making occur. It demonstrates that the genetic algorithm formalism provides a computational construct to carry out this learning. The process is exemplified using a shape grammar for a beam section. A new shape grammar is learned that produces a new state space for the problem. This new state space has improved characteristics.

This article presents an assessment of four management systems to expose the essential characteristics of each management system, including planning techniques, problem representation, concurrency, and flexibility. The experimental part of the research shows that existing management systems can be used to attack a variety of problems. The authors conclude that flexible planning systems are quite beneficial since they can be used to solve a variety of design problems by making small modifications in the definition of their tools.

The status of research methodology employed by studies on the application of AI techniques to solving problems in engineering design, analysis, and manufacturing is poor. There may be many reasons for this status, including: unfortunate heritage from AI, poor educational system, and researchers’ sloppiness. Understanding this status is a prerequisite for improvement. The study of research methodology can promote such understanding, but, most importantly, it can assist in improving the situation. Concepts from the philosophy of science are introduced, and models of world views of science are built on them. These world views are combined with research heuristics or research perspectives and criteria for evaluating research to create a layered model of research methodology. This layered model can serve to organize and facilitate a better understanding of future studies of research methodologies. Many of the issues involved in the study of AI and AIEDAM research methodology using this layered model are discussed.

The application of truth maintenance techniques to the component selection phase of the engineering design process is described. The aims of the research include the selection of components that most closely match a set of requirements which include fluid characteristics, economic constraints, and company policies or preferences. In addition, we consider the storage and retrieval of requirement sets and corresponding component selections. The benefits of maintaining temporal knowledge are twofold: to enhance performance through reuse when new requirement sets are similar to previous sets, and to re-assess selections in the light of changes over time, for example, to the set of available components, to company policy, or, perhaps, to national safety regulations.

We propose a temporal truth maintenance system (TTMS) to support this management of selection knowledge over time.

The use of machine learning techniques requires the formulation of a learning problem in a particular domain. The application of machine learning techniques in a design domain requires the consideration of the representation of the learned design knowledge, that is, a target representation, as well as the content and form of the training data, or design examples. This paper examines the use of a target representation of design concepts and the application, adaptation, or generation of machine learning techniques to generate design concepts from design examples. The examples are taken from the domain of bridge design. The primary machine learning paradigm considered is concept formation.

Spreadsheets are difficult to use in applications, where only incomplete or inexact data (e.g., intervals) are available-a typical situation in various design and planning tasks. It can be argued that this is due to two fundamental shortcomings of the computational paradigm underlying spreadsheets. First, the distinction between input and output cells has to be fixed before computations. Second, cells may have only exact values. As a result, spread-sheets support the user only with primitive iterative problem solving schemes based on trial-and-error methods. A constraint-based computational paradigm for next generation interval spreadsheets is presented. The scheme makes it possible to exploit incomplete/inexact data (intervals), and it can support problem solving in a top-down fashion. Current spreadsheets constitute a special case of the more general interval constraint spreadsheets proposed.

An approach for transforming the order of magnitude relation between two variables into an algebraic equality or inequality constraint is provided. In order to derive the order of magnitude relation between any two variables, a nonlinear optimization problem is solved for the minimum and maximum values of the ratio between the two variables, subject to two classes of constraints. The first class of constraints corresponds to the quantitative model and the second class of constraints corresponds to the qualitative model. The optimization approach is shown to provide more precise inferences as compared to the conventional constraint satisfaction approaches. Moreover, this approach provides a crucial step in developing unified frameworks that allow the incorporation of qualitative information at various levels of abstraction into numerical frameworks used for reasoning with quantitative models.

Less than a decade ago it seemed that a new paradigm of engineering–called computer-aided engineering (CAE) – was emerging. This emergence was driven in part by the success of computer support for the tasks of engineering analysis and in part by a new understanding of how computational ideas largely rooted in artificial intelligence (AI) could perhaps improve the practice of engineering, especially in the area of design synthesis. However, while this “revolution” has failed to take root or flourish as a separate discipline, it has spawned research that is very different from traditional engineering research. To the extent that such CAE research is different in style and paradigm, it must also be evaluated according to different metrics. Some of the metrics that can be used are suggested, and some of the evaluation issues that remain as open questions are pointed out.

Computational abstraction of engineering design leads to an elegant theory defining (1) the process of design as an abstract model of computability, the Turing machine; (2) the artifacts of design as enumerated strings from a (possibly multidimensional) grammar; and (3) design specifications or constraints as formal state changes that govern string enumeration. Using this theory, it is shown that engineering design is a computable function. A computational methodology based on the theory is then developed that can be described as a form follows function design paradigm.

Perspectives on research methodology in the field of “AI and design” are discussed. This perspective is based on a view of a “science of design” focusing on methods of design and on characteristics of design tasks that affect what methods are relevant for a given task. Two methodological issues are discussed: the need to try applying a design method on multiple tasks and domains, and the need to work with collaborators who are experts in the task domain of each research system you build.

One method for making analogies is to access and instantiate abstract domain principles, and one method for acquiring knowledge of abstract principles is to discover them from experience. We view generalization over experiences in the absence of any prior knowledge of the target principle as the task of hypothesis formation, a subtask of discovery. Also, we view the use of the hypothesized principles for analogical design as the task of hypothesis testing, another subtask of discovery. In this paper, we focus on discovery of physical principles by generalization over design experiences in the domain of physical devices. Some important issues in generalization from experiences are what to generalize from an experience, how far to generalize, and what methods to use. We represent a reasoner's comprehension of specific designs in the form of structure-behavior-function (SBF) models. An SBF model provides a functional and causal explanation of the working of a device. We represent domain principles as device-independent behavior-function (BF) models. We show that (1) the function of a device determines what to generalize from its SBF model, (2) the SBF model itself suggests how far to generalize, and (3) the typology of functions indicates what method to use.

Human designers often adopt strategies from previous similar cases to guide their search in new design tasks. We have developed an approach for automated design strategy capture and reuse. That approach has been implemented in DDIS, a prototype structural design system that uses a blackboard framework to integrate case-based and domain-based reasoning. Plans, goals, and critical constraints from user-selected previous cases are combined with case-independent reasoning to solve underconstrained parametric structural design problems. This article presents a detailed example of design strategy recording and reuse in base plate design for electrical transmission poles.

A multifaceted approach to evaluating expert systems is overviewed. This approach has three facets: a technical facet, for “looking inside the black box”; an empirical facet, for assessing the system’s impact on performance; and a subjective facet, for obtaining users’ judgments about the system. Such an approach is required to test the system against the different types of criteria of interest to sponsors and users and is consistent with evolving lifecycle paradigms. Moreover, such an approach leads to the application of different evaluation methods to answer different types of evaluation questions. Different evaluation methods for each facet are overviewed.

Design knowledge is continually refined and expanded through experience. This research is concerned with design knowledge expressed as constraints. A simple learning mechanism simulates an expert designer's ability to incrementally adjust her knowledge when presented with slightly new problems. In response to unsatisfied expectations during the design process the system will examine its general knowledge about the design artifact, discover some relevant constraining knowledge, and convert that knowledge into a design constraint for future use. This process, referred to as constraint inheritance, should automatically improve the problem-solving performance.

An approach to providing computational support for concurrent design is discussed in the context of an industrial cable harness design problem. Key issues include the development of an architecture that supports collaboration among specialists, the development of hierarchical representations that capture different characteristics of the design, and the decomposition of tasks to achieve a trade-off between efficiency and robustness. An architecture is presented in which the main design tasks are supported by agents – asynchronous and semiautonomous modules that automate routine design tasks and provide specialized interfaces for working on particular aspects of the design. The agent communication and coordination mechanisms permit members of an engineering team to work concurrently, at different levels of detail and on different versions of the design. The design is represented hierarchically, with detailed models maintained by the participating agents. Abstractions of the detailed models, called “agent model images,” are shared with other agents. In conjunction with the architecture and design representations, issues pertaining to the exchange of information among different views of the design, management of dependencies and constraints, and propagation of design changes are discussed.

This article poses the notion that it is possible and desirable to formalize and apply design critiques in a specialized framework. It describes GENCRIT, (GENeric CRItiquing Tool), one such framework for design critiquing. The article starts by highlighting the role of critics in the design process. It then goes on to bring out the need for a critic building tool, viz. that of aiding in the rapid development of multiple critics. GENCRIT combines knowledge-based techniques and a multifactor decision making model to develop an integrated approach to evaluation that encompasses a wide range of designs. Critics developed using GENCRIT evaluate candidate designs based on the critiquing knowledge provided by experts, give justifications for the evaluation, and suggest improvements. The working of GENCRIT is illustrated with two examples: a constructibility critic for reinforced concrete buildings and a bridge design critic.