The AAAI Special Interest Group on AI in Manufacturing (SIGMAN) organization represents a wide array of researchers addressing fundamental scientific problems that arise in engineering design and manufacturing. The human endeavor that is design and manufacturing is rich in basic problems for researchers in Artificial Intelligence. From process planning researchers who aim to augment the abilities of human process engineers, to those who employ agents and distributed AI to model and control the factory floor—AI has found one of its most fertile and challenging proving grounds in the manufacture of products.

This paper reports on the results of an analysis of existing ontological systems to determine which is most appropriate for the manufacturing domain. In particular, this involved the exploration of efforts that are studying both the uses of ontologies in the general sense and those that are using ontologies for domain-specific purposes. Eleven ontological systems were analyzed and, using a set of analysis criteria, it was determined that the Cyc (Cyc is a registered trademark of Cycorp Inc.) system was most appropriate for modeling concepts in the manufacturing domain. After the analysis is described, examples are given to show how manufacturing concepts could be modeled in the Cyc system. This work is part of a larger project whose objective is to move closer to the ultimate goal of seamless manufacturing systems integration using the principle behind ontological engineering to unambiguously define domain-specific concepts. The output of this work will be a taxonomy of manufacturing terms and concepts along with formal definitions of exactly what each of those terms and concepts mean and how they interrelate.

This paper proposes interval constraint networks and interval propagation techniques for automatic tolerance design. To develop an intelligent automatic tolerance design mechanism, the following interrelated problems are addressed: (1) representing the relationships between the tolerances of the entities, attributes, and the functional requirements of a design component by a constraint network, (2) developing tolerance propagation techniques in the network that will satisfy the constraints between the entities, attributes and functional requirement. For the purpose of tolerance design, we propose new definitions for consistency of constraints and consistency of a constraint network. New forward and backward propagation techniques achieve worst-case tolerance analysis and synthesis, respectively. A tank design problem illustrates the use of the procedures for tolerance analysis and synthesis. The discussion and examples compare the new constraint network and propagation techniques with previous works.

This paper describes the process planning techniques we developed for use in an Integrated Product and Process Design (IPPD) tool for the design and manufacture of microwave transmit/receive modules. Given a collection of data about the design of a microwave module, the IPPD tool uses a combination of AI planning and OR trade-off analysis to produce a collection of alternative designs and alternative process plans that have Pareto optimal values for manufacturing and purchasing lead time, process yield, cost, and number of suppliers. The IPPD tool provides facilities to enable the user to generate and examine these Pareto optimal alternatives in real time, in order to provide immediate feedback on how to modify the design to improve its cost and productivity.

In this work, we describe an approach to automated manufacturing feature extraction and operations planning that combines the two processes to benefit both. When viewed in planning terms, if feature extraction is the process of identifying planning goals, and operations planning is the the process of selecting and sequencing planning operators, then what we are doing can be viewed as combining the normally separate processes of finding goals and developing instantiated operators to satisfy them. Thus, we call this approach Ebgoc (Effector-Based Goal and Operator Construction), and we have implemented it in a computer program called Mediator. Effectors are the physical equipment, such as tools and machines, that are used to transform the initial materials into the desired end product (goal). We say that this approach is effector-based (rather than feature-based) because we do not recognize a fixed set of features, but instead we geometrically derive the set of shapes that can be machined with the currently available set of effectors. Thus, each shop can customize Mediator so that it identifies machinable volumes appropriate for the resource in that specific shop. This approach gives Mediator several important properties. It can 1) effectively handle feature interactions (e.g., volumetric intersections), 2) be customized to produce features appropriate to specific shop resources, 3) identify areas that the current resources cannot machine, and 4) handle nonstandard, user-defined tooling.

Considering manufacturing expertise during the early stages of design can be of great benefit. Such information can greatly improve not only the quality of a design, but it can also ensure the generation of an easily manufactured design. This, in turn, can lower the final cost of the designed product. By evaluating how easily an evolving conceptual design can be made, potential hazards can be avoided before any detailed design efforts commence. The conceptual manufacturing planning requisite to such an evaluation is the focus of this paper. A domain-independent strategy for conceptual manufacturing planning is presented. A task-structure analysis of this strategy shows its domain independence. A specific implementation of this strategy for polymer composites manufacturing planning (Socharis) is discussed. The high-level implementation details of Socharis are presented as instantiations of the conceptual manufacturing planning strategy. Finally, the validity of the planning strategy and the utility of Socharis are assessed.

The use of agent-based systems may offer significant advantages over centralized hierarchical manufacturing systems, including distribution of control/processing and data as well as adaptable automated or semiautomated problem solving. However, simply applying any agent-based paradigm to manufacturing domains may not be enough to address the various demands of these domains. Manufacturing domains require agent-based problem solving to be flexible and tolerant of faulty information (e.g., due to faulty sensors), equipments, and communication links. These issues are addressed by research on Sensible Agents, capable of 1) Dynamic Adaptive Autonomy and 2) explicit Perspective Modeling. Dynamic Adaptive Autonomy allows Sensible Agents to change problem-solving roles during system operation to address dynamic factory floor conditions, while Perspective Models internal to a respective Sensible Agent serve to maintain knowledge about the agent itself, other agents, and the environment. The application of the Sensible Agent paradigm to a shop-floor control is demonstrated.