Abstraction As a Configuration Design Methodology

Author(s):  
Gary L. Snavely ◽  
Panos Y. Papalambros

Abstract Configuration design can be thought of as a process of generating artifacts by assembling pre-defined components. This paper introduces a method for reducing the size of configuration problems by abstracting components to higher levels of abstraction. At higher abstraction levels, less important detail is temporarily ignored, and each component represents a family of lower-level components. Configuration is then performed at the highest level, explicitly enumerating all configurations at that level. Any complete configuration at the highest level is recursively instantiated to lower levels. At the same time, any incomplete configuration at the highest level is eliminated, thereby eliminating all possible lower-level instantiations of that configuration. In this manner, all configurations of components at the lowest level of abstraction are implicitly enumerated.

2012 ◽  
Vol 134 (4) ◽  
Author(s):  
Benjamin W. Caldwell ◽  
Gregory M. Mocko

Function modeling is often used in the conceptual design phase as an approach to capture a form-independent purpose of a product. Previous research uses a repository of reverse-engineered function models to support conceptual-based design tools, such as similarity and design-by-analogy. These models, however, are created at a different level of abstraction than models created in conceptual design for new products. In this paper, a set of pruning rules is developed to generate an abstract, conceptual-level model from a reverse-engineered function model. The conceptual-level models are compared to two additional levels of abstraction that are available in a design repository. The abstract models developed through the pruning rules are tested using a similarity metric to understand their usefulness in conceptual design. The similarity of 128 products is computed using the Functional Basis controlled vocabulary and a matrix-based similarity metric at each level of abstraction. A matrix-based clustering algorithm is then applied to the similarity results to identify groups of similar products. A subset of these products is studied to further compare the three levels of abstraction and to validate the pruning rules. It is shown that the pruning rules are able to convert reverse-engineered function models to conceptual-level models, better supporting design-by-analogy, a conceptual-stage design activity.


Author(s):  
Linda C. Schmidt ◽  
Jonathan Cagan

Abstract A computational approach to design that integrates conceptual design, configuration design, and catalog component selection tasks overcomes some of the barriers to successful design automation. FFREADA is a design generation and optimization algorithm featuring hierarchical ordering of grammar based-design generation processes at different levels of abstraction. FFREADA is used to design hand-held, power drills and to develop an appropriate objective function for design optimization. The drill grammar expresses a vast space of design states that are not limited to any particular functional architecture or component configuration. (The algorithm’s optimization runs operate in a space which exceeds 20249 designs.) Good drill designs, those with values within 1% of the optimal solution, are found in minutes by sampling less than 0.15% of the design states. Optimal configurations are found for drills with three different torque requirements.


Author(s):  
Jesse D. Peplinski ◽  
Patrick N. Koch ◽  
Janet K. Allen ◽  
Farrokh Mistree

Abstract How can design for manufacture be implemented very early on a design time-line, for example, when only the function is known? Our response is embodied in an approach to design for manufacture at the function level of abstraction based on the notion of design using available assets. In this paper we focus on the solution scheme and computer implementation of our approach to design for manufacture. Our solution scheme takes the form of a Heuristic Selection Decision Support Problem, and our computer tool is called FLAME: the Function Level of Abstraction Manufacturability Evaluator. We use this tool to identify, evaluate and select potential manufacturing alternatives for products modeled at the function level of abstraction. We illustrate some of its uses by exploring the selection of manufacturing processes and materials for a component from a design of an aircraft evacuation system, although our focus is on the method rather than on the results per se.


Author(s):  
Huy Tran ◽  
Ta’id Holmes ◽  
Uwe Zdun ◽  
Schahram Dustdar

This chapter introduces a view-based, model-driven approach for process-driven, service-oriented architectures. A typical business process consists of numerous tangled concerns, such as the process control flow, service invocations, fault handling, transactions, and so on. Our view-based approach separates these concerns into a number of tailored perspectives at different abstraction levels. On the one hand, the separation of process concerns helps reducing the complexity of process development by breaking a business process into appropriate architectural views. On the other hand, the separation of levels of abstraction offers appropriately adapted views to stakeholders, and therefore, helps quickly re-act to changes at the business level and at the technical level as well. Our approach is realized as a model-driven tool-chain for business process development.


Author(s):  
FEDERICO BERGENTI ◽  
AGOSTINO POGGI

Software engineering relies on the possibility of describing a system at different levels of abstraction. Agent-oriented software engineering introduces a new level of abstraction, that we called agent level, to allow the architect modelling a system in terms of interacting agents. This level of abstraction is not supported by an accepted set of tools and notations yet, even if a number of proposals are available. This paper introduces: (i) An UML-based notation capable of modelling a system at the agent level and (ii) A development framework, called ParADE, exploiting such a notation. The notation we propose is formalized in terms of a UML profile and it supports the realisation of artefacts modelling two basic concepts of the agent level, i.e., the architecture of the multi-agent system and the ontology followed by agents. The choice of formalising our notation in terms of a UML profile allows using it with any off-the-shelf CASE tool. The ParADE framework takes advantage of this choice by providing a code generator capable of producing skeletons of FIPA-compliant agents from XMI files of agent-oriented models. The developer is requested to complete the generated skeletons exploiting the services that ParADE and the underlying agent platform provide.


2020 ◽  
Vol 58 (1) ◽  
pp. 51-61
Author(s):  
Satoshi Gunji ◽  
Kotaro Tonoike ◽  
Jean-Baptiste Clavel ◽  
Isabelle Duhamel

Author(s):  
W. Ernst Eder

The engineering design methodology of Pahl and Beitz is good in the detailed stages, but needs enhancement in the early stages of conceptualizing and embodiment-in-principle. The concept of ‘functions’ has been enhanced by Hubka and colleagues. A ‘functional basis’ (Hirtz et al) has improved the definitions of ‘flows’ and ‘functions’, their work does not go far enough to provide a basis for conceptualizing. ‘Affordances’ (Maier and Fadel) are covered by full use of systematic conceptualizing of design engineering solutions. The Pahl-Beitz model and method of ‘decomposition of functions’, ‘physics’, and components is contrasted with the Hubka models of a transformation system, TrfS, its constituents, structures, properties life cycle, etc., and their use as method for design engineering by searching for alternative embodiments at each of these levels of abstraction. These steps are illustrated in (to date) 21 case examples published between 1976 and 2012, several of them in the CEEA conferences and their predecessors.


2021 ◽  
Vol 6 (1) ◽  
pp. 54-61
Author(s):  
Willy C. Shih

Many new technologies are complex and embody high levels of technical sophistication, and applying them should require significant knowledge and experience. Yet, the rapid adoption and incorporation of these technologies into other innovations seems inconsistent with the expertise needed to make them work. In this paper, we propose increasing levels of abstraction as a strategy for speeding the adoption of new technologies. Higher-level abstractions package complexity in ways that makes them easier to understand and recombine, and they decrease the resources needed by firms to deploy sophisticated technical know-how. Increasing the level of abstraction is a way to push forward the innovative frontier by making such difficult-to-use technologies readily accessible to other innovators. Although this framing has been used in engineering and software development to describe modular encapsulation and cumulative innovation, we propose its use in the management literature to describe more broadly the uptake of new technologies and their facile recombination. This framing casts a different light on cumulative innovation and exposes new managerial questions to explore


Sign in / Sign up

Export Citation Format

Share Document