Philosophical grounding and computational formalization for practice based engineering knowledge

https://doi.org/10.1016/j.knosys.2006.06.002Get rights and content

Abstract

Michael Polanyi’s idea of tacit knowing and Martin Heidegger’s concept of pre-theoretical shared practice are presented as providing a strong rationale for the notion of practice based knowledge. Artificial Intelligence (AI) approaches such as Artificial Neural Networks (ANN), Case Based Reasoning (CBR) and Grounded Theory (with Interval Probability Theory) are able to model these philosophical concepts related to practice based knowledge. The AI techniques appropriate for modeling Polanyi’s and Heidegger’s ideas should be founded more on a connectionist rather than a cognitivist paradigm. Examples from engineering practice are used to demonstrate how the above techniques can capture, structure and make available such knowledge to practitioners.

Section snippets

Background and objectives

Theoretical knowledge has been prized in academic institutions at least since the scientific revolution. The philosophical underpinnings for this, in the form of privileging the intellectual over the practical, have come from Descartes, but deeper roots lie in Plato himself. In engineering, this focusing on theoretical knowledge has caused a gap between academic training and professional practice, as the latter often calls for practitioner judgement and experience [10]. At the same time, many

Michael Polanyi – Tacit knowing

One of Polanyi’s main contributions to epistemology was the idea of tacit knowing; one of his books is titled “The Tacit Dimension” [25]. A key aspect of tacit knowing was that it attended from particulars to a whole. Polanyi used the example of recognizing a face to illustrate this – we use our subsidiary awareness of the features in order to achieve focal awareness of the face [25]. The important thing was that the particulars should not be focused on, but “seen through”, like a pair of

Martin Heidegger – pre-theoretical shared practices

One of the main thrusts of Heidegger’s philosophy is the primacy of practice, or rather practices that we are socialized into, prior to any theoretical understanding [19]. Heidegger approached the question of being from what he called “the human way of being”. He did this because humans were the only beings who were concerned about their own being. He used the term Da-sein to denote this being. In addition to meaning “the human way of being”, this hyphenated German word can also mean

Categories of practice based knowledge

Both Polanyi and Heidegger are good advocates for the importance, and indeed the primacy of practice based knowledge. How then can this knowledge be formalized and categorized? We have said before that AI can in fact provide a formalization for practice based knowledge. Within AI, Minsky [21] has distinguished between cognitivist and connectionist approaches to knowledge.

The cognitive approach is epitomized by expert systems [3]. Here, the knowledge is made explicit, by eliciting it from an

Modelling tacit knowledge: construction bid decisions

Let us now look at some examples where AI techniques have been used to capture and process practice based knowledge. Consider the modelling of tacit knowledge. There are many areas in engineering that are characterized by such knowledge, none more clearly than bidding for construction projects, decisions for which have been described as being made “on the basis of intuition derived from a mixture of gut feelings, experience and guesses” [1]. The language is very reminiscent of Polanyi. There is

Modelling shared practice: layout design

The objective of this study was to explore the potential for using Artificial Neural Networks (ANNs) and Case Based Reasoning (CBR) for suggesting column spacing and sizing in multistory buildings, based on historical examples [12]. Column spacing and sizing are part of preliminary design, and often based on “engineering judgment”, which can be considered an aspect of shared practice. Data were obtained for a total of 45 existing buildings from different design offices; hence the data genuinely

Modelling horizontal knowledge: vulnerability of buildings to bomb blast

The examples given above are based essentially on historical knowledge that has been structured into various fields (e.g. factors that affect bid mark-up on the one hand and grid spacing on the other). Such structuring could be construed as imposing a cognitive framework on the practice based knowledge and hence departing somewhat from the connectionist paradigm. The interactions between the fields however are genuinely connectionist, in that there are no cognitive rules that combine evidence

Discussion

At this stage we discuss an issue each from the philosophical and computational aspects in this paper and seek to further clarify their inter-relatedness. The first issue has to do with a comparison of Polanyi and Heidegger. On the one hand, they are poles apart. Heidegger is a very nihilistic philosopher who advocated a “hermeneutic of suspicion”, while Polanyi sought to restore a fiduciary (or faith like) framework for the practice of science. Polanyi’s focus is on epistemology, and hence he

Conclusions

  • 1.

    We have seen that the epistemology of Michael Polanyi and the ontology of Martin Heidegger provide a significant intellectual basis for the notion of practice based knowledge.

  • 2.

    We have demonstrated that Artificial Intelligence (AI) techniques such as Artificial Neural Networks (ANN) and Case Based Reasoning (CBR) can model philosophical concepts such as tacit knowing (Polanyi) and shared practice (Heidegger).

  • 3.

    The juxtaposition of the above philosophical grounding and computational formalizations

References (33)

  • F. Baird et al.

    An ethnographic study of engineering design teams at Rolls Royce Aerospace

    Design Studies

    (2000)
  • I. Ahmad

    Decision support system for modelling bid/no-bid decision problem

    ASCE Journal of Construction Engineering and Management

    (1990)
  • Building Expert Systems (1983) (Eds.) Hayes-Roth, F., Waterman, D.A. and Lenat, D.B., Addison-Wesley,...
  • R. Chandratilake et al.

    Identifying vulnerability of buildings to blast events using Grounded Theory, in: Proceedings of the 10th Annual Symposium on Research for Industry

    (2004)
  • R.D. Coyne

    Design reasoning without explanations

    AI Magazine

    (1990)
  • W.C. Cui et al.

    Interval probability theory for evidential support

    International Journal of Intelligent Systems

    (1990)
  • John Dewey
  • John Dewey
  • W.P.S. Dias

    Reflective Practice, Artificial Intelligence and Engineering Design: Common Trends and Inter-relationships, Artificial Intelligence in Engineering Design

    Analysis and Manufacture (AIEDAM)

    (2002)
  • W.P.S. Dias, D.I. Blockley, Reflective practice in engineering design, ICE Proceedings on Civil Engineering 108 (4)...
  • W.P.S. Dias, S.R. Chandratilake, Assessing vulnerability of buildings to blast using Interval Probability Theory, in:...
  • W.P.S. Dias, U.A. Padukka, AI Techniques for Preliminary Design decisions on column spacing and sizing, in: Proceedings...
  • W.P.S. Dias et al.

    Artificial neural networks for construction bid decisions

    Civil Engineering Systems

    (1996)
  • W.P.S. Dias et al.

    Dimensions of order in engineering design organizations

    Design Studies

    (2002)
  • H.L. Dreyfus

    Husserl, Heidegger and Modern existentialism

  • B. Glaser et al.

    The Discovery of Grounded Theory: Strategies for Qualitative Research

    (1967)
    • View full text
    Copyright © 2006 Elsevier B.V. All rights reserved.