Manufacturing process planning is a difficult problem with a prohibitively large search space. It is normally tackled by decomposing goal objects into features, and then sequencing features to obtain a plan. This paper investigates an alternative approach. The capabilities of a manufacturing process are represented by a formal language of shape, in which sentences correspond to manufacturable objects. The language is interpreted to describe process plans corresponding to the shape generation, complete with cost estimates. A macro layer that describes single operations of the machine is implemented on top of the formal language. The space it describes is searched by the generative simulated annealing algorithm, a stochastic search technique based on simulated annealing. Plans that are close to the optimum are generated in reasonable time.

When model-based planning systems are scaled up to deal with full-sized industrial projects, the resulting complexity in the project-specific model and production plan can create serious problems, not only in dealing with such complexity computationally, but also in user-acceptance. In the model-based planning system described in this paper, activities are dynamically generated, inherently at the detailed level of individual physical components. However, it is possible to intelligently group together collections of components which would be common to realistic work packages, and hence schedule on the basis of virtual components existing within an abstraction hierarchy. This paper describes a technique of project planning within an integrated design/planning system, which exploits fundamental knowledge of engineered systems and provides powerful and flexible planning functionality.

Previous research on qualitative reasoning about shape and fit laid the foundations to determine whether two objects fit together. Continued investigation has refined the theory and has produced a functioning implementation. This paper describes extensions to the theory and the details of the implementation.

The reasoning process has been divided into five layers: grouping, topology, orientation, matching, and confirmation. The grouping layer clusters features such as cubes or cylinders into groups for each surface of an object. The topology layer recognizes patterns formed by the groups on each surface, and describes the pattern in terms of topological structures. The orientation layer selects promising surfaces from the two objects and attempts to align the two surfaces. If the orientation layer aligns the topological structures on the two surfaces, the matching layer tries to pair the features within the topological structures. The confirmation layer inspects paired features to determine whether the surfaces are compatible. If the surfaces are compatible, then the two objects qualitatively fit together.

Arguing that design is a social process, we expand the meaning of modeling and analysis to include all activities facilitating continual refinement and criticism of the design requirements, process and solutions. We do not assume any a priori methods for modeling or analysis; rather, we provide a framework and an approach to study designers and give them whatever modeling and analysis capabilities they choose. Our approach is the basis for a support tool, n–dim, currently under development.

Conflicts are likely to arise among participants in a collaborative design process as the inevitable outgrowth of the differing perspectives and viewpoints involved. The opportunities for conflict are magnified if many perspectives are brought to bear on a common artifact early in the design process, as in concurrent engineering or integrated engineering. Design advice tools can assist in the process of resolving these conflicts by making critiques and suggestions conveniently available to design participants, and by offering a fair means of evaluating and comparing suggested alternatives for compromise solution. In previous work we introduced a protocol based on notions of economic utility by which design advice systems can recognize conflict and mediate negotiation fairly. This protocol allowed design teams to express the desire to maximize or minimize the values of design parameters over totally ordered bounded domains of values, such as real numeric intervals. In this paper we extend this approach by allowing expressed preferences of design teams to be qualitative as well as quantitative, by allowing teams to express interest in parameters before they actually come into existence, and by relaxing many other of the earlier restrictions on the ways teams may express their preferences.

Computational simulation of physical systems generally requires human experts to set up a simulation, run it, evaluate the quality of the simulation output, and repeatedly invoke the simulator with modified input until a satisfactory output quality is achieved. This reliance on human experts makes use of simulators by other programs difficult and unreliable, though invocation of simulators by other programs is critical for important tasks such as automated engineering design optimization. Presented is a framework for constructing intelligent controllers for computational simulators that can automatically detect a wide variety of problems that lead to low-quality simulation output, using a set of evaluation methods based on knowledge of physics and numerical analysis stored in a data/knowledgebase of models and simulations. An experimental implementation of this framework in an intelligent automated controller for a widely used computational fluid dynamics simulator is described.

Development of expert systems involves knowledge acquisition that can be supported by applying machine learning techniques. The basic idea of using decision-tree induction in process optimization and development of the domain model of electrochemical machining (ECM) is presented. How decision-tree induction is used to build and refine the knowledge base of the process is also discussed.

The idea of developing an intelligent supervisory system with a learning component [Intelligent MAnufacturing FOreman (IMAFO)] that is already implemented is briefly introduced. The results of applying IMAFO for analyzing data from the ECM process are presented. How the domain model of the process (electrochemical machining) is built from the initial known information, and how the results of decision-tree induction can be used to optimize the model of the process and further refine the knowledge base are shown. Two examples are given to demonstrate how new rules (to be included in the knowledge base of an expert system) are generated from the rules induced by IMAFO. The procedure to refine these types of rules is also explained.

It is now becoming increasingly common for products to be designed via an interdisciplinary approach, particularly when an engineering enterprise adopts the concurrent engineering approach. Essentially, this implies that several heterogeneous sources of knowledge are simultaneously involved during the design phase of the product. The benefits often cited as a result of such an approach are increased quality and innovation, decreased time-to-market, and lower manufacturing costs. However, the approach also accentuates the problem of how to handle and manage conflicts, which frequently arise due to several factors as discussed in this paper. This paper also describes a system called Schemebuilder, which is an integrated design workbench aimed at supporting the conceptual and embodiment phases of interdisciplinary systems design, and its mechanisms for handling and resolving conflicts.

Most design tasks involve the management of conflict. Conflict arises when contradictory requirements are imposed upon characteristics of artifacts, upon the process of their creation and/or upon their intended use. Even individual design requires trade-offs because of competing design criteria, such as functionality, safety, cost, and social acceptance. The ability of designers to avoid or minimize conflict through judicious tradeoffs, careful negotiations and other methods become their most valuable skills.

Research in computer-aided design/engineering (CAD/E) has focused on enhancing the capability of computer systems in a design environment, and this work has continued in this trend by illustrating the use of the Dempster-Shafer theory to expand the computer’s role in a CAD/E environment. An expert system was created using Dempster-Shafer methods that effectively modeled the professional judgment of a skilled tribologist in the selection of rolling element bearings. A qualitative and symbolic approach was used, but access to simple quantitative models was provided to the expert system shell. Although there has been significant discussion in the literature regarding modification/improvement of the Dempster-Shafer theory, Shafer’s theories were found adequate in all respects for replicating the expert’s judgment. However, an understanding of the basic theory is required for interpreting the results.

The design of even the simplest product requires thousands of decisions. Yet few of these decisions are supported with methods on paper or on computers. Is this because engineering design decisions do not need support or is it because techniques have yet to be developed that are usable on a wide basis? In considering this question a wide range of decision problem characteristics need to be addressed. In engineering design some decisions are made by individuals, others by teams – some are about the product and others about the processes that support the product – some are based on complete, consistent, quantitative data and others on sparse, conflicting, qualitative discussions. To address the reasons why so little support is used and the characteristics of potentially useful decision support tools, a taxonomy of decision characteristics is proposed. This taxonomy is used to classify current techniques and to define the requirements for an ideal engineering design decision support system.