Software productivity has been steadily increasing over the last 30 years, but not enough to close the gap between the demands placed on the software industry and what the state of the practice can deliver [22,39]; nothing short of an order of magnitude increase in productivity will extricate the software industry from its perennial crisis [39,67]. Several decades of intensive research in software engineering and artificial intelligence left few alternatives but sofware reuse as the (only) realistic approach to bring about the gains of productivity and quality that the software industry needs. In this paper, we discuss the implications of reuse on the production, with an emphasis on the technical challenges. Software reuse involves building software that is reusable by design, and building with reusable software. Software reuse includes reusing both the products of previous software projects, and the processes deployed to produce them, leading to a wide spectrum of reuse approaches, from the building blocks (reusing products) approach on one hand, to the generative or reusable processor (reusing processes) on the other [68]. We discuss the implications of such appproaches on the organization, control, and method of software development and discuss proposed models for their economic analysis. Software reuse benefits from methodologies and tools to: 1) build more readily reusable software, and 2) locate, evaluate, and tailor reusable software, the latter being critical for the building blocks approach. Both sets of issues are discussed in this paper, with a focus on application generators and object-oriented development for the first, and a thorough discussion of retrieval techniques for software components, component composition (or bottom-up design) and transformational systems for the second. We conclude by highlighting areas that, in our opinion, are worthy of further investigation.

United States Government. Approved for public release; distribution unlimited.

Analogical reasoning has a long history in artificial intelligence research, primarily because of its promise for Ike acquisition unit effective use of knowledge. Defined as a representational mapping from a known "source " domain into a novel "target" domain, analogy provides a basic mechanism for effectively connecting a reasoner's past and present experience. Using a four-component process model of analogical reasoning, this paper reviews sixteen computational studies of analogy. These studies are organized chronologically within broadly defined task domains of automated deduction, problem solving and planning, natural language comprehension, and machine learning. Drawing on these detailed reviews, a comparative analysis of diverse contributions to basic analogy processes identifies recurrent problems for studies of analogy and common approaches to their solution. The paper concludes by arguing that computational studies of analogy are in a slate of adolescence: looking to more mature research areas in artificial intelligence for robust accounts of basic reasoning processes and drawing upon a long tradition of research in other disciplines.

The paradigms of transformational planning, case-based planning, and plan debugging all involve a process known as plan adaptation --- modifying or repairing an old plan so it solves a new problem. In this paper we provide a domain-independent algorithm for plan adaptation, demonstrate that it is sound, complete, and systematic, and compare it to other adaptation algorithms in the literature. Our approach is based on a view of planning as searching a graph of partial plans. Generative planning starts at the graph's root and moves from node to node using planrefinement operators. In planning by adaptation, a library plan---an arbitrary node in the plan graph---is the starting point for the search, and the plan-adaptation algorithm can apply both the same refinement operators available to a generative planner and can also retract constraints and steps from the plan. Our algorithm's completeness ensures that the adaptation algorithm will eventually search the entire graph and its systemat...

Patdex is an expert system which carries out case-based reasoning for the fault diagnosis of complex machines. It is integrated in the Moltke workbench for technical diagnosis, which was developed at the university of Kaiserslautern over the past years, Moltke contains other parts as well, in particular a model-based approach; in Patdex where essentially the heuristic features are located. The use of cases also plays an important role for knowledge acquisition. In this paper we describe Patdex from a principal point of view and embed its main concepts into a theoretical framework 1 General Considerations Patdex 1 is an expert system which carries out case-based reasoning for the fault diagnosis of complex machines. It is integrated in the Moltke workbench 2 for technical diagnosis, which was developed at the university of Kaiserslautern over the past years (cf. e.g. [4, 5, 23]), Moltke contains other parts as well (cf. e.g. [16]), in particular a model-based approach (cf. [21, ...

. Structured domains are characterized by the fact that there is an intrinsic dependency between certain key elements in the domain. Considering these dependencies leads to better performance of the planning systems, and it is an important factor for determining the relevance of the cases stored in a case-base. However, testing for cases that meet these dependencies, decreases the performance of case-based planning, as other criterions need also to be consider for determining this relevance. We present a domain-independent architecture that explicitly represents these dependencies so that retrieving relevant cases is ensured without negatively affecting the performance of the case-based planning process. 1 Introduction Reusing previous problem solving experience has proven to speed-up planning systems. Problem solving experience can be stored in generalized (Minton, 1988) or abstracted (Bergmann & Wilke, 1995) form, or it can be stored as cases (Veloso, 1994; Ihrig & Kambhampati, 1994...

An inportant form of inductive learning is inventing a meaningful classification of given objects or events. This chapter extends the authors' previous work on this problem based on conceptual clustering, i.e., grouping objects into conceptually simple classes. In contrast to the past work, the new method deals with classifying objects represented by structural descriptions rather than sequences of attribute values. These descriptions are expressed in Annotated Predicate Calculus (A_PC), which is a typed predicate logic calculus with additional operators.

ion and respecialization is a general structural adaptation technique that abstracts the piece of the retrieved solution, and respecializes it later. Respecialization results in applying other specializations of the abstractions to the current situation. Thus, it results in analogical problem solving. This technique is used in PLEXUS (Alterman, 1986) as a second step in adaptation process (first, a null adaptation is used, and only if something fails, the system replans it using abstraction and respecialization). A similar approach is used in PERSUADER (Sycara, 1987). 4. Critic-based adaptation is a structural adaptation based on using critics to debug almost correct solutions (Simmons, 1988; Hammond, 1989a; Gonzalez and Laureano-Ortiz, 1992). A critic checks if a particular combination of features can cause a problem in a plan. If such a feature is found, a specific repair strategy is applied for repair. In CHEF, several criticbased adaptation rules are used, e.g., deletion of unneces...

06801. Thanks go to Professor R. S. Michalski tor supervising this project, and contributing