Texts
- The Problem of Learning
- Problemistics Courseware
- Corso su Problemistica
- Resources Management
- Manuale/Intellettuale
- Campagna/Città
Problemistics - Problémistique - Problemistica
The Art & Craft of Problem Dealing
Forecasting
Definition (Union of International Associations)
Exploratory and normative (Erich Jantsch)
Planning (Erich Jantsch)
Classification (M. S. Makower and E. Williamson)
Classification (Michael Carley)
Classification (C. H. Waddington)
Future research (Union of International Associations)
Relevance tree (C. H. Waddington)
Forecasting: relevance tree (Erich Jantsch)
Trend extrapolation (Erich Jantsch)
Trend extrapolation (M. S. Makower and E. Williamson)
Trend extrapolation (George Chadwick)
Delphi (Erich Jantsch)
Logistic curves (Erich Jantsch)
Envelope curves (Erich Jantsch)
Scenario writing (Erich Jantsch)
Scenario writing (C. H. Waddington)
Morphological analysis (Erich Jantsch)
Predictability (Eric Duckworth)
Requirements (Herbert A. Simon)
[1986, Second Edition] Encyclopaedia of World Problems and Human Potential, Union of International Associations eds. K. G. Saur, München
"A forecasting is a probabilistic statement, on a relatively high confidence level, about the future (as contrasted with a prediction, which is a statement on an absolute confidence level about the future). A distinction is made between technological forecasting and social forecasting." (KC0547)
[1972] Erich Jantsch, Technological Planning and Social Futures, Associated Business Programmes, London
"The development of a methodology for technological forecasting took a decisive turn when a clear distinction was drawn between
- exploratory (opportunity-oriented)
- normative (need-oriented)
approaches." (p. 63)
“Exploratory forecasting, which starts from today’s level of knowledge and explores future feasibilities and probabilities, has to match normative forecasting, which implies the delineation of goals of the future and their translation into missions and tasks for scientific and technological development.” (p. 48)
[1972] Erich Jantsch, Technological Planning and Social Futures, Associated Business Programmes, London
"Forecasting is an integral part of the planning process which, in turn, unfolds in the cybernetic process of rational creative action. To engage this process fully, we have to view it at three levels, linked by feedback interaction between them:
- policies (what ought we to do?)
- strategies (what can we do?)
- operations or tactics (what will happen if we take a specific course of action?)
This vertical integration of planning and action brings human values and norms into play in an important way.
The horizontal integration of planning and action aims at encompassing planning for complex multivariate systems." (A Framework for Thought and Action, p. 9)
[1975] M. S. Makower & E. Williamson, Operational Research, Hodder & Stoughton, London
“Methods of forecasting may be of the following kinds:
1. Statistical analysis of past data of the variable being forecast.
2. Statistical analysis of data of other variables which are shown to be related to the one of interest.
3. Intuition, inside information, etc.”
(Chapter 4, p. 62)
[1980] Michael Carley, Rational Techniques in Policy Analysis, Heinemann Educational Books, London
Three types of future research have been identified:
- Trend extrapolation. "Such techniques implicitly assume that factors which influenced past phenomena will continue to have the same or similar effect in the future, and that the direction and rate of this effect can be estimated." (p. 146) (Techniques: Growth or S-curves; Quantified analogies; Substitution curves. see p. 147)
- Qualitative. "These techniques are qualitative in that they rely on human judgements and criticism gathered by the asking of questions of opinion and value."(p. 147) (Techniques: Scenario writing; Delphi, Cross-impact analysis; Gaming; Counterfactual Analysis)
- Dynamic Modelling. "... the questions used in the dynamic modelling process had to reflect (i) proven physical relationships (ii) tested postulated relationships from the social sciences or (iii) qualitative judgements as to relationships." (p. 152)
[1977] C.H. Waddington, Tools for Thought, Jonathan Cape, London
- Exploratory. "In exploratory forecasting, the forecaster attempts to explore by examination of and extrapolation from the present, the possible future states of any particular situation." "Exploratory forecasts ... are useful in mapping out the choices which are available to planners and decision-makers, given existing technology, population, income, social patterns, etc. and their likely future development." (pp. 198-200)
- Normative. "Whereas exploratory forecasts are concerned with what future changes and developments are possible, normative forecasts are concerned with identifying long-term goals and desirable futures and with working out the steps by which this desirable future state may be brought about." (p. 215) "Realistic normative forecasts should mesh with the framework of possibilities predicted by exploratory forecasts." (p. 216)
[1986, Second Edition] Encyclopaedia of World Problems and Human Potential, Union of International Associations eds. K. G. Saur, München
"1. A broad range of interlinked activities which include:
(a) conjectural, speculative and imaginative descriptions of the human future;
(b) exploratory forecasts based on methodological extrapolation of past and present developments into the future; and
(c) prescriptions, namely normatively oriented future projections in which explicit value assertions and choices are made about how a specific future may be viewed.
2. Any serious organized attempt to devise concepts and methods which can be used to conjecture intelligently about the human future." (KC 0450)
[1977] C.H. Waddington, Tools for Thought, Jonathan Cape, London
"Relevance trees are hierarchically ordered networks of objectives, the main purpose of which is an aid to decision-making in complex situations in which it may not be obvious where to allocate one’s resources. A vertical relevance tree relates a long-term objective to the alternative actions that should be taken in order to achieve that objective." (Chapter 11, p. 216)
[1972] Erich Jantsch, Technological Planning and Social Futures, Associated Business Programmes, London
“Relevance trees, or vertical relevance analysis, constitute another ‘orderly way of looking at things’, but from the other end, starting with broad objectives and goals and attempting to set up hierarchical relationships for all conceivable contributions to them.” “Relevance trees constitute ... the only systematic approach in use linking the entire range of levels.” (p. 87)
“Horizontal relevance analysis is nothing else but plotting a number of important systemic relationships of a non-hierarchical character.” (p. 90)
[1972] Erich Jantsch, Technological Planning and Social Futures, Associated Business Programmes, London
“Trend extrapolation is the oldest forecasting technique. The extrapolation of trends over time usually assumes that the specific constellation of forces influencing growth over the recent past, will continue its influence in the future, possibly with some marginal modification.”
(Chapter 6, p. 83)
[1975] M. S. Makower & E. Williamson, Operational Research, Hodder & Stoughton, London
"One way of expressing the problem of forecasting is the question: how can one select the appropriate form of regularity (with its associated patterns of deviations) on which to base one’s forecast?" (p. 64)
- (a) No trend (ex. 5,5,5,5,)
- (b) Linear trend (ex. 5,6,7,8,9,...)
- (c) Exponential growth (ex. 4,8,16,32,64,...)
- (d) Exponential decay (ex. 64,32,16,8,4,...)
- (e) Cyclical (ex. 15,8,15,8,15,8,...) [seasonal if referred to one year span]
- (f) Linear trend and cyclical (ex. 4,6,8,10,4,6,8,10,...)
(from Chapter 4, pp. 64-65)
[1971] George Chadwick, A Systems View of Planning, Pergamon Press, Oxford
Exploratory techniques: trend extrapolation
"Five types of trend curves have been distinguished.
- the first type is where increase is linear, becoming flatter in slope where some limit is approached;
- the second case is of exponential increase without flattening, giving a straight line when the logarithm of the factor is plotted over time;
- the third class is of S-shaped curved (the logistic is such a curve);
- a fourth class is that of double exponential, or even steeper growth with subsequent flattening;
- the fifth is that of slow exponential increase followed by sudden rapid increase, with eventual flattening (e.g. the growth of world population)."
"An important aspect of the use of such curves ... lies in the potential existence of an ‘envelope curve’ which contains a family of several lower order curves of the same kind." (Chapter 7, pp. 162-165)
[1972] Erich Jantsch, Technological Planning and Social Futures, Associated Business Programmes, London
"The Delphi technique ... is essentially a refinement of the original ‘brainstorming’ technique. The difference is that in seeking the views of many experts, procedures are used to make these experts sharpen their own thinking and to prevent them from exchanging opinions with one another, Delphi panel members never meet, but communicate with a ‘control centre’ (never with other panel members), usually by a succession of ‘rounds’ of letters." (Chapter 6, p. 82)
[1972] Erich Jantsch, Technological Planning and Social Futures, Associated Business Programmes, London
"S-curves, or logistic curves, are the graphic expression of growth under restraint, in particular under the influence of absolute or practical limits, not many of which are usually known in advance (which makes trend extrapolation of technological growth risky)." (Chapter 6, p. 84)
[1972] Erich Jantsch, Technological Planning and Social Futures, Associated Business Programmes, London
"The basic idea is that a broad technological capability - for example speed attained by man - may be extrapolated in the same way in which capabilities of a specific technology may be extrapolated. But that technology capability is now seen as being not restricted by the maturing of a specific technology, but subject to being pushed further by a succession of different technologies. The result is usually a ‘big S’ (the envelope curve) riding on a series of small ‘S’s’." (Chapter 6, pp. 85-86)
[1972] Erich Jantsch, Technological Planning and Social Futures, Associated Business Programmes, London
"Scenario writing ... is essentially an attempt to view and combine various trends in a systemic way. Typically, small single-dimensional steps are taken (for example, in technological, economic, or politic developments), the changes implied for the scenario are mapped, and decisions which have to be made (or may be made) due to these changes are then systematically explored. In this way, various sequences of small logical steps into the future are explored to obtain a feeling for the long-range systemic consequences. The primary purpose is not to predict the future, but systematically to explore branching points dependent upon critical choices. Scenario writing ... is a formalised approach to mapping alternative courses of action into the future."
(Chapter 6, pp. 86-87)
[1977] C.H. Waddington, Tools for Thought, Jonathan Cape, London
"Scenario writing is a technique in which the writer attempts to set up a logical series of events in order to show how, starting from the present (or any other given situation), a future situation may evolve step by step. The purpose is not to predict the future but to generate plausible future situations." (Chapter 11, p. 203)
[1972] Erich Jantsch, Technological Planning and Social Futures, Associated Business Programmes, London
"Morphological analysis, in the words of its pioneer Fritz Zwicky who applied it as early as 1942, is ‘an orderly way of looking at things’ and so achieving ‘a systematic perspective over all possible solutions to a given large-scale problem’. This is an exploratory approach which attempts to break up the problem into its basic parameters, and then conceive of as many variations [combinations?] of each parameter as possible."
(Chapter 6, p. 87)
[1967] Eric Duckworth, A Guide to Operational Research, Methuen & Co., London, First Published 1962
"Life in this world is possible, and interesting, because we can neither predict the future nor is it unpredictable." (Conclusion, p. 135)
[1988, First edition 1969] Herbert A. Simon, The Sciences of the Artificial, M.I.T. Press, Cambridge Massachusetts
"Good predictions have two requisites that are often hard to come by. First they require either a theoretical understanding of the phenomena to be predicted, as a basis for the prediction model, or phenomena that are sufficiently regular that they can simply be extrapolated." "The second requisite for prediction is having reliable data about the initial conditions - the starting point from which the extrapolation is to be made. Systems vary in the extent to which their paths are sensitive to small changes in initial conditions." (Chapter 6, p. 170)