posted 1 Jul 2000 in Volume 3 Issue 10
Problem solving through enacted
While objective knowledge is essential to problem-solving, so too is the creativity and know-how of individual workers. Based on a survey on the information-seeking activities of a group of 27 senior engineers, Martin Ward explores the flow of knowledge that contributes to the solution of given problems in an engineering consultancy firm.
This contribution to our thinking about knowledge management starts from the idea that knowledge, be it explicit or tacit, is dead and useless until it is enacted by the engineer at the critical junctures of the work-flow. Enacted knowledge is the interaction between objectively valid knowledge, and the consciousness of the engineer, expressed in action: Action by the whole person, not a numerical unit, an imperfect machine or a detached mind.
For the past 80 years, Ricardo’s consultancy work with engines and transmissions of all kinds has contributed to the design of most of the vehicles on our roads, as well as a host off off-highway machines, locomotives, pumps, ships and tractors all over the world. Transportation is more than an economic necessity and a convenience; it gives us the power of movement that is at the heart of our modern sense of being. This power depends on the work of engineers like those described in this paper.
This paper uses findings from a survey of the information-seeking activities of 27 senior engineers which was carried out by the author in 1996-7. Although many changes have taken place in information management and engineering since then, the basic direction of the findings is still valid. A full report of the survey is available separately(1).
In practical terms, the life of the engineer at Ricardo is a never-ending sequence of problems, mostly set by clients. The success of Ricardo’s staff in dealing with this work-flow is reflected in Ricardo’s survival, current financial success, and expansion.
The literature on problem solving is immense, as reference to the titles of current books in print will testify. The catalogue of the West Sussex County Libraries refers the user to the whole of ‘logic’. A typical definition from a management text widely used at Ricardo is that “problem solving is a question of gathering, structuring and using information”(2). Only two engineers in the survey disagreed with this.
One of the constituents of successful problem solving, therefore, is the provision of clear, objective, accurate and reliable information, delivered on time. A perceived gap in information or knowledge may be part of the problem. For paper studies of engines and components, or for answers to technical questions submitted by users of Ricardo’s Technical Support Service, it may be virtually the whole problem.
The survey confirmed a number of sources from which this information is taken by the engineers:
1. Personal files of notes, information, past papers, and test results, held in the offices of individuals. One respondent claimed that these were the second most important source.
2. Personally-owned text and reference books. The engineers varied in their possession and use of these.
3. Departmental files, books and databases. They represent an important halfway house between the engineers’ personalised collections and the formal collection in the Ricardo library. The Design Office, for historical reasons, has the most extensive ‘library’ of this kind. Some departments have virtually nothing.
4. Records of previous work. These are divided between one and three, above, and the library. The convenience of having internal records close at hand explains and justifies the duplication that occurred here.
5. The Ricardo library, managed by Ricardo’s Information Services Department (ISD). It has possibly the largest collection of powertrain literature in the world, which is indexed by the POWERLINK database. This is based on DB/TextWorks software provided by Inmagic. At the time of the survey, STATUS E4 software was in use. Both programmes afford free-text access to many thousands of bibliographical references. POWERLINK is available both within Ricardo and to its clients, for a certain fee.
6. Other sources. These included industry manuals, brochures, papers from external conferences, letters and privately-held periodicals.
In previous user studies, these sources have been described as ‘formal’(3). In knowledge management terms, they correspond to explicit knowledge, recorded on paper or its substitutes. ‘Formal’ knowledge is represented in Popper’s epistemology by World Three(4), in contrast to Worlds One (physical objects) and Two (consciousness). World Three, though collectively produced, exists independently of any individual consciousness (World Two) with its tendencies towards subjectivity, prejudice and error, as an objective system of knowledge. It is knowledge of this kind that problem solving needs. Magee, following Popper(5), claims that this is the only knowledge worth having. For the engineers, World Three knowledge is set within the rational, objective, epistemological framework of the disciplines of engineering.
However, the survey also showed that formal knowledge, while necessary for problem solving, is not sufficient in practice. As well as the sources mentioned above, the engineers relied heavily on what previous surveys have called ‘informal’ sources. These were cited as:
7. Memory. One respondent claimed this was the resource of first resort.
8. Other people. All the respondents relied on the judgement, direction, and knowledge of their peers.
9. Other informal sources. They included clients, manufacturers, suppliers and leisure contacts.
In knowledge management terms, these sources correspond to tacit knowledge. For Popper and McGee they correspond to World Two. McGee virtually disavows the validity of such knowledge.
Subjective and objective views of knowledge
What McGee’s view of knowledge really requires is a mechanical, infallible machine to solve the problems. Although many of the engineers thought that parts of their job could be taken over by expert systems or other mechanical devices, most in the survey believed that their individual judgement would still be required. They naturally had a personal interest in saying this, but a look at the way they solved problems rather supports their view.
The view of this writer, and of others concerned with knowledge management, is that people are not merely imperfect machines(6,7,8), much though they have been treated in this way by past management theories(9). A fact in a book is useless until it has been assimilated in the engineer’s mind, and brought through action to the critical point of the work flow. Useful action requires, in its final stage, tacit knowledge. Without consciousness, knowledge is dead and useless.
What kind of consciousness is required? An information specialist can ‘know’ the rated speed of an engine, but lacks the surrounding technical judgement which the engineer alone possesses. Engineers have internalised World Three through a long and challenging process of acculturation. Their knowledge is part of them. It constitutes most of their social role as engineers.
Like everyone else, they have prejudices and biases and are prone to subjective error. But they can distinguish within their own minds the difference between these personal attributes and the objective knowledge they possess; this is what being a professional means.
This view is supported by Michael Polanyi’s book Personal Knowledge(10), which argues for knowledge as a personal possession. He cites the doctor who learns after long training to read an X-ray as an example of this. His book gives the case for recognising this aspect of the mind as ‘responsible consciousness’ (my phrase).
There must also be another level of consciousness at work, which allows the engineer to extract objective tacit knowledge from his mind and apply it to the specific requirements of the work-flow at a particular time. This is called ‘know-how’, and stands on the subjective side of a spectrum between objective knowledge and total subjectivity. It is indispensable for solving problems.
Yet another requirement for problem solving, as reported in the survey, is a form of the very subjectivity that McGee so denigrates. The engineers in the survey persistently reported the role of insight in their work. One of them said that problem solving first involves becoming aware of a problem, then a period of exploration to get a ‘feel’ for the situation. This is followed by the construction of a possible solution. It is important to recognise – and this takes engineering judgement – when the right solution has presented itself. The process involves a ‘leap in the dark’. Another engineer asserted the roles of know-how, intuition, judgement and insight. Asked how they arrived at insight, six engineers said they used rational methods such as logic, while eight used non-rational means, such as a flash of insight, frequently given to them when they were engaged in some other activity.
It is pertinent at this point to note that one of the classic studies of problem solving is Arthur Koestler’s The Act of Creation(11). Creativity, an elusive, subjective talent, is nearby in the solving of engineering problems.
This has moved the argument about problem solving rather a long way from the text book definition with which we began, which described it as the mechanical application of information.
Enacted knowledge is the interaction between essential, objective knowledge, and the consciousness of an actor, expressed in his action. An action is the work of a person, not an imperfect machine or a detached mind. Enacted knowledge is necessary for problem solving.
Let us look in greater detail at the individual traits of the engineers in the survey. They had a variety of problem solving styles. Indeed, the variety of their replies – not one question of the 72 asked elicited a unanimous response – is surprising in a population so sociologically homogeneous. They came from a wide range of disciplines, from diesel engineering (6 respondents), computer aided engineering (5), design (3), research (3), design analysis (2), control (2), gasoline engines (2) and noise (1), among others. The average number of years they had worked at Ricardo was 14, demonstrating a steady career commitment. For 11 of them, Ricardo was their first and only affiliation. They were all male, and all principal engineers, forming the backbone of Ricardo’s collective expertise.
The questionnaire invited the respondents to classify themselves as convergent or divergent thinkers (there were no resources to test this response objectively). Eight claimed to be convergent thinkers, and eight divergent. Eleven rejected the alternatives, claiming to be able to adopt either style according to the type of problem facing them.
Eight said they espoused thoroughness and method, eschewing the use of intuition and short cuts, and eight habitually used the latter approaches. This time there were seven rejectors, who, again, claimed to be able to be able to use either approach as a strategy, depending on the problem.
The respondents in the survey differed in their perception of their role as engineers. Sixteen accepted the administrative work involved, while 11 were unhappy with it. Some of the older respondents did not see computer literacy as essential to their engineering role, with consequences for their attitude to ISD’s POWERLINK database. Two engineers expressed strongly pro-information and pro-library attitudes; others were less enthusiastic, which probably also affected their attitude to POWERLINK.
Autonomy in problem solving
Individuals solving problems need autonomy. On the one hand, the free-text nature of POWERLINK, and the ability to consult it in the personalised environment of their own offices, reinforced the autonomy of engineers as information seekers. Unfortunately, the user-unfriendly nature of STATUS E4’s command language diminished their autonomy, as an objective test of their facility with it showed. Several respondents also complained about the opaque and unfriendly nature of the book arrangement and its classification.
Autonomy in problem solving has to be balanced by control, if the unproductive aspects of subjectivity are to be kept in check. Ricardo’s social system itself rewards adherence to corporate objectives, both formally through promotions and bonuses, and informally through peer pressure. In addition, the quality assurance system rigorously enforces agreed norms which include a system by which problems should be approached and solved. Some respondents expressed a slightly rebellious attitude towards the system; others were responsible for organising it.
Whatever their feelings, all the engineers had long ago internalised within themselves a large measure of the institutions controlling their work, ensuring that negative aspects of subjectivity, such as memory loss, prejudice and error, were held in check within their own minds, as one would expect from professionals.
Overall constraints of time and cost (time is cost in a consultancy like this) have a pervasive effect on information-seeking and problem solving. This cultural condition surfaced in the survey when respondents preferred single-shot to interactive library searches, and complained that there was not enough time to browse in the library. The quality assurance system imposes a rigid timetable on the sequence of events in solving problems.
A society of mind
Problem solving is carried out by individuals, but not on their own. The work of an inventor connected with Ricardo was studied during the survey. It revealed creativity in problem solving, but also the lonely quest which that often involved. The case of the engineers at Ricardo is different; they rely very heavily on the opinions, knowledge and judgement of their peers. Ricardo has an open, communicative culture. The survey noted the existence of secret projects, with confidential information gathering and recording, and the occasional individual (never named), who hoarded knowledge for personal, political gain. These were exceptions to a very open society.
The survey revealed the existence of knowledge clubs, made of informal groups of engineers, transcending departmental borders and disciplines, who habitually relied on each other, particularly those with good memories, for advice. The 27 respondents named 96 individuals between them, forming on an average a cluster of six colleagues each.
The concept of the knowledge club is similar to that of the technological gatekeeper(12). But there are basic differences: The latter is an exceptional individual, the former were a group, and a group of quite ordinary engineers.
Patterns of problem-solving and learning
Problem solving is partly cyclic, the application of known methods to different, but not new, problems. On the other hand, it can be progressive and incremental, adding to the individual and corporate knowledge base, by assimilation of related, or accommodation of totally new, knowledge. This distinction of learning styles is drawn from Jean Piaget’s work(13).
Only two engineers in the survey said they did not deal with new problems. But even a cyclic pattern of problem solving deepens and sharpens knowledge and skills in encountering a new situation; this might be called incidental learning. On the other hand, six engineers said they dealt with some new things, while 13 said they dealt with new things all the time. Four applied existing methods to new problems. The gaining of radically new knowledge in this way could be described as conspicuous learning.
The results of problem solving
The lessons learned from solving problems were generally written up in a report for the client, enlarging the latter’s knowledge. The report, when filed in the library, would deepen Ricardo’s explicit knowledge, while the tacit knowledge of the participating engineers would be accessible to peers. Frequently, solutions found their way into published papers, enlarging the knowledge of the engineering community as a whole.
The survey invited the engineers to distinguish between formal and informal knowledge-sources – between explicit and tacit knowledge – and to evaluate them. This is probably an unnatural distinction in the blurry conditions of real life, when tacit and explicit knowledge constantly criss-cross, and duplicate each other. Nevertheless, some engineers found the library to be the top source of information. They drew, from browsing there, some of the spark and inspiration that they found in the tacit knowledge of their peers. Others, on the contrary, found the latter the most helpful source. One engineer in a preliminary survey to help draw up the questionnaire said: “Ten minutes with the right person is worth a week in the library.” Another concluded that: “Jack is my library.”
People share subjectivity; books do not. In conversation, information comes in neat, customised bundles, pre-sorted and tagged with engineering judgements. Both the questions and the replies are open to negotiation, unlike texts. To obtain the same value from a document might required a laborious extra stage of tagging and assimilating.
The content of the engineers’ tacit knowledge, and that of the explicit knowledge in the library overlap. They clearly do so, since many of the documents there were written by Ricardo’s engineers.
People are quicker. Explicit knowledge cannot issue in useful action until it has been absorbed as tacit knowledge. Reading can be defined as the process of converting explicit to tacit knowledge. It is then ready to become enacted knowledge.
But explicit knowledge, such as the Ricardo library provides, is essential, as the engineers’ replies made clear. No-one can remember everything, and opinions of peers have to be backed by hard data before they can be tried out on clients. Only documents, paper knowledge, can supply this. There is therefore no conflict between tacit and explicit knowledge; rather, they are complementary, with the proviso that the final stage belongs to tacit knowledge, in the form of enacted knowledge.
In this paper, we have taken the traditional opposition between informal and formal information sources, and applied them to their true context in terms of tacit and explicit knowledge. We at Ricardo are not scientists or academics. Knowledge is not the final goal, useful action is – action at the critical junctures of the work-flow.
Let Arie de Gues’s The Living Company13 suggest the wider context of our work. In it, he argues for a view of the company as a community of individuals with aspirations, loyalties and professional commitments, working together as an organic unity. Although profit is an essential outcome of their work, based on high-quality service to clients, the role of people, as people, and not as numbers, comes clearly through the survey, and the thoughts to which it has given rise. KM
1. In process of publication. Send an e-mail to MLWard@ricardo.com
2. Straker, David. A Toolbook for Quality Improvement and Problem Solving, (London, Prentice Hall, 1995)
3. Roberston, Andrew, ‘Behaviour patterns of scientists and engineers in information seeking for problem solving’, ASLIB Proceedings, (October 1974, Vol. 26, No. 10, p385)
4. Popper, Karl, Objective Knowledge: An Evolutionary Approach, (Oxford University Press, 1975)
5. Magee, Bryan; Popper, Fontana, Modern Masters, (1975)
6. Nonaka, Ikujiro and Hiro Takeguchi, The knowledge-Creating Company: How Japanese Companies Create the Dynamics of Innovation, (Oxford University Press, 1995)
7. Senge, Peter M, The Fifth Discipline: The Art and Practice of the Learning Organization, (New York and London, Doubleday Currency, 1990)
8. Leadbetter, Charles, Living on Thin Air: The New Economy, (London, Viking, 1999)
9. Taylor, Frederick Winslow, The Principles of Scientific Management, (1911)
10. Polanyi, Michael, Personal Knowledge: Towards a Post-Critical Philosophy, (London, Routledge & Kegan Paul, 1958)
11. Koestler, Arthur, The Act of Creation, (Hutchinson, 1964)
12. Allen, Thomas, Managing the Flow of Technology, (New York, Harper & Row, 1977)
13. Discussed in Arie de Guess, The Living Company: Growth, Learning and Longevity in Business, (London & Naperville, Nicholas Brearley Publishing Ltd, 1999)
Martin Ward is an information specialist working for the Information Services Department of Ricardo Consulting Engineers. He can be contacted at: MLWard@ricardo.com