5 METHODOLOGY .1 Scope
5.6 Methodological Lnterpretation of the Methanol Experiments
Armed with new insight from the above sections on methodology, we may now go back and re-examine the results of t h e simulation experiments described in Section 4.3.
Each of t h e three cases considered in t h a t section illustrates a particu- lar aspect of t h e methodology:
Case 1: Investment-oriented comparative study of methanol technologies Case 2: Flexibility of t h e PDA with respect to the critical resources, i.e., tra-
deoff analysis
Case 3: Quest for concordance between available technologies and available resources, with a range of estimates of resource availability.
In t h e methanol study we have a specific PDA based on a s e t of technologies which offer alternative routes from various raw materials to methanol. This naturally brings the analysis down to the level of a comparative study of tech- nologies. If t h e decision maker is "investment sensitive" we have a returns-to-investment problem; this is illustrated by Case 1. In. t h e same way i t would be possible t o base the comparative analysis on the efficiency of energy use, raw material consumption, or manpower utilization of t h e various techno1ogi.e~.
Let us now take a n o t h e r look a t t h e t h r e e cases described in Section 4 . 3 given level of investment. These technologies and their capacity utilization a r e given explicitly in Table 7. This information opens t h e way for f u r t h e r described above, t h e performance of a particular cornbination of technologies can be m e a s u r e d by a single value. Thus t h e Attainable Performance Area with respect t o critical resources, and may therefore be seen as a n analysis of tradeoffs between t h e critical resources.
*Note that since only new technologies are considered the unconstrained best present (UHP) state does not exist.
**Note that the energy considered here is that actually used in the technological processes (steam, electricity, etc.).
C a s e 3. w e s t for c o n c o r d a n c e
We have emphasized t h a t learning is a n i m p o r t a n t f a c t o r in t h e decision process t h a t we define as a q u e s t for concordance between available r e s o u r c e s a n d technologies. Cases 1 a n d 2 r e p r e s e n t learning steps in t h i s process: t h e y help t h e decision m a k e r t o acquire some understanding of t h e properties of t h e PDA in t e r m s of critical resources. Some of t h e data used t o facilitate learning a n d t o provide a basis for t h e final decision should be t r e a t e d with caution: t h e r e a r e obvious u n c e r t a i n t i e s associated with
"apparent" availabilities of resources a n d estimated operating r e q u i r e m e n t s of hypothetical plants.
The critical resources e s t i m a t e d m a y include: a m o u n t of raw m a t e r i a l s (gas, coal, e t c . ) available for use, t h e a m o u n t of capital available for invest- m e n t , a n d t h e desired levels of various processes or effects. All of t h e s e esti- m a t e s a r e subject t o analysis a n d may be obtained from various sources
-
t h e y generally r e s u l t from studies carried o u t by experts.
The m o s t n a t u r a l way of formulating t h e e s t i m a t e s is t o express t h e m in t e r m s of ranges, as discussed earlier. Algorithms using this idea of admissible ranges for critical resources were used in Case 3, which also illustrates o t h e r i m p o r t a n t aspects of our methodology.
The essence of t h e q u e s t for concordarice is t o find a compromise between t h e estimated availability of resources, t h e aspirations of t h e deci- sion m a k e r , and t h e capabilities of t h e technologies. The PDA is basically a s t r u c t u r e t h a t trarlsforms inputs i n t o outputs. The desired effects of t h i s transformation a r e expressed by t h e criteria, while t h e e n v i r o n m e n t in which i t m u s t take place is specified by t h e constraints. The purpose of t h e exercise is t o obtain answers t o th.e following questions:
1. What is t h e b e s t s t r u c t u r e of t h e PDA from t h e point of view of t h e vari- ous criteria?
2. Can t h e expectations of t h e decision maker be m e t ?
The following situations may a r i s e in connection with t h e second of t h e s e questions:
- The decision m a k e r overestimates t h e potential of t h e s y s t e m : his aspirations a r e too high a n d c a n n o t be achieved;
-
The decision m a k e r u n d e r e s t i m a t e s t h e potential of t h e s y s t e m : his aspirations can be fulfilled a n d even exceeded.In t h e first case t h e c o m p u t e r informs t h e decision m a k e r of t h e feasible solution closest to his expectations. The decision m a k e r t h e n e i t h e r accepts t h i s solution, revises his aspirations, or gives u p .
In t h e second case t h e computer informs t h e decision m a k e r of t h e b e s t possible r e s u l t , which actually exceeds his expectations This does n o t m e a n t h a t analysis of t h e e s t i m a t e s should now stop: t h e reason for t h e difference between e s t i m a t e s a n d results should be found. If t h e s o l u t ~ o n i s b e t t e r t h a n expected, t h i s provldes s o m e information about t h e estlmatlon m e t h o d used.
The m e t h o d is n o t necessarily wrong, i t may simply be incompatible with t h e s y s t e m .
There is actually a third possibility (and one to which t h e interactive process should ultimately lead.): t h a t the best solution lies within t h e estimated ranges. The safest course of action is to accept a solution which is equidistant from t h e limits imposed by the ranges (i.e., on the skeleton of the ADS
-
s e e Section 5.2). Even in this case the decision maker should carry out f u r t h e r analysis of the results, since these only confirm t h a t t h e PDA is con- sistent with t h e estimates of resources.The situations d e s c r i b e d above a r e illustrated by t h e results given in Table 10. In experiment 1 3 t h e possibilities have been overestimated, while in experiment 14 they have been underestimated. By contrast, experiments 11 and 12 a r e good examples of concordance. Comparison of these four experi- m e n t s is very instructive since i t iIlustrates the effects of subjective prefer- ences or external factors (expressed by the admissible ranges). Recall t h a t t h e results given in t h e table a r e "safe" in t e r m s of equidistance from the limits of t h e ranges: comparison of experiments 11 and 12 shows t h a t t h e higher expectation of efficiency in experiment 12 is fulfilled only a t t h e cost of additional investment.
The above considerations are summarized and illustrated in Figure 18.
This shows how the preferences of the decision maker (in t e r m s of t h e ADS and the "safety" represented by i t s skeleton - see Figure 19) a r e transformed to yield t h e Pareto-optimal s t r u c t u r e of t h e PDA.
Efficiency
+ max (109 m . u . 1 7
FlGURE 18 Graphical interpretation of sample experiments from Table 10. Here
-
@-
is the skeleton of t h e ADS for experiment number n,+
---+
represents the approximation of t h e Pareto s e t ,+
represents t h e solution obtained i n experiment number n a n d ( E / I ) , is t h e corresponding efficiency/investment ratio.Q2 -+ rnin FIGURE 19 Gra hical interpretation of t h e admissible d e m a n d s e t and i t s skeleton.
P :
Here Qiu a n d Qi I epresent t h e upper a n d lower bounds of t h e admissible ranges (i = 1, 2).
The sequence of experiments shown in Figure 18 (see also Table 10) illus- trates how the decision maker learns and adapts his preferences to attain the concordant solution. These experiments not only yield the values of objective functions but also identify the best s t r u c t u r e of the PDA in t e r m s of the cri- teria, specifying particular technologies and their parameters, production levels, amounts of resources, etc. (see the Appendix).
It is also worth looking a t what happens when the preferences (aspira- tions) of the decision maker are incompatible witti the properties of the PDA.
Recall in particular t h a t on p. 34 we pointed out t h e contradiction between t h e required minimum methanol supply level and actual methanol produc- tion in experiments 11, 12, and 14 (Table 10). In these cases there is no s t r u c t u r e t h a t simultaneously (i) belongs to t h e PDA, (ii) is the "best" in t e r m s of t h e criteria, and (iii) meets the demand for methanol.
Table 12 provides another illustration of the problem of concordance, but in this case emphasis is placed on the availability of raw materials r a t h e r than on investment, which is simply given a fixed value. The critical raw material here is natural gas.
The type of data obtained from a typical experiment is illustrated in the Appendix.
6 CONCLUSIONS
This paper represents t h e first and possibly t h e most important step i n our research on the problem of developing new sources of chemical feedstocks through industrial s t r u c t u r a l change.
The PDA concept provides a very useful way of structuring the problem, confirming expectations based on i t s use i n earlier work (generating a development program for t h e chemical industry). This has given us a consid- erable degree of confidence in t h e PDA model. The methodology developed so far makes use of and develops our earlier concept of a quest for concordance between available resources and available technologies. Although f u r t h e r methodological work remains to be done, t h e theoretical basis of the approach appears t o be sound, and t h e results obtained so far seem very encouraging.
The computer software developed i n parallel with t h e methodology has also proved itself in practice, b u t will be modified in the future t o accommo- date changes in methodology o r improvements suggested by further experi- ments. The data collected for t h e methanol study have proved very illuminat- ing and a r e now also being used for other purposes.
Our results a r e in general agreement with papers dealing with the use of fossil resources for energy supply (see Hafele e t al. 1982, Sassin 1982, Rogner
1982). We note with satisfaction t h a t energy researchers are moving percepti- bly closer t o our area - t h e emerging common ground may be called energo- chemical processing of fossil resources. Since t h e r e is an explicit com- plementarity between these approaches, we expect to cooperate closely with t h e relevant IIASA projects ( P a t t e r n s of Economic Structural Change and Industrial Adjustment, and particularly Energy Development and Invest- ments) in the future.
For t h e dual reasons of clarity and limited space, this report does not include many of our findings. These a r e concerned with various problems based on prices and their evaluation using sensitivity analysis and postop- timal analysis, and will be published separately. Much important mathemati- cal background has also been omitted.
We shall continue with this general line of research in the future, emphasizing not only tools and their methodological implications but also case studies of specific PDAs. We intend to pay special attention to the ener- gochemical processj.ng of lignite.
We a r e convinced by t h e results of our study that. a temporary drop in oil prices should not dater decision makers and researchers from t h e pursuit of new sources of raw materials and energy, and new ways of deploying t h e m . The world cannot afford t o waste any t i m e in this field. Should the present inefficient patterns of resource utilization continue, t h i s waste of time will only be translated into a waste of resources.
ACKNOWLEDGEMENTS
We a r e grateful t o all of t h e people who have helped u s a t various stages of t h e research reported h e r e . However, we would like t o express our especial gratitude t o Professor H. Gorecki for his patient guidance through t h e theoretical problems t h a t we encountered; t o Professor A. Wierzbicki, who was always very supportive, both as leader of t h e System and Decision Sciences Program a t IIASA and as our c o n s u l t a n t in t h e a r e a of multiobjective optimi- zation a n d interactive methodology; and t o Janusz Kindler, Leader of t h e Resources and Environment Area a t IIASA, whose common-sense approach and help could always be relied upon.
H. Rogner pointed out several contentious points in a n earlier draft, and we a r e also very grateful t o t h e anonymous referees for t h e i r constructive c o m m e n t s , which were very helpful when revising t h e paper.
We should n o t forget our Editor, Helen Gasking, who patiently fought h e r way through our terrible m a n u s c r i p t a n d clarified i t greatly.
Any remaining e r r o r s a r e , of course, ours.
APPENDIX
In Table A. 1 we p r e s e n t t h e complete s e t of data obtained from one of t h e experiments described in Section 4.3. We have selected experiment 15 (see Tables 12 and 13) as our example, b u t t h e s a m e type of information is avail- able from all experiments 1-18.
It is clear t h a t some of t h e information is concerned with t h e design of installations belonging to t h e PDA under investigation. These data become very i m p o r t a n t once a particular case (an experiment) h a s been chosen for f u r t h e r analysis, and could be obtained from a n initial study carried out by a design office. These data m a y help t h e decision maker to reach t h e n e x t stage in his investment decision or may lead h i m t o reconsider t h e m a t t e r com- pletely.
TABLE A 1 Set of data obtained from a typical experiment (based on experiment 15 from Tables 12 and 13).
Parameter Value Parameter Value
General B y - p r o d u c t s
Efficiency 8.35 x
loe
m.u. Higher alcohols 17 xlo3
tonsEnergy consumption 1791 X
lo3
t.c.e. Phenol 13 xlo3
t o n sI n v e s t m e n t 20 x
l o 9
m . u . Fuel oil 88 x l o 3 t o n sLabor 409 m e n Sulfur 21 x l o 3 t o n s
Methanol p r o d u c t i o n 945 x l o 3 t o n s
W a s t e p r o d u c t s
Raw m a t e r i a l s C O ~ 629 x 1
o3
Nm3Natural g a s 400 x l o 3 Nm3 Waste w a t e r 2724 x l o 3 m3
Coal 1862 x l o 3 t o n s Ash 216 x l o 3 t o n s
Carbide g a s 102 x
lo3
Nm3ZnO c a t a l y s t 62 t o n s C o n s t r u c t i o n d a t a
Reforming c a t a l y s t 31 t o n s Weight of c o n c r e t e 95,064 t o n s Shift-conversion c a t a l y s t 44 t o n s Weight of s t e e l s t r u c t u r e 12,363 t o n s Methanation c a t a l y s t 5 9 t o n s Weight of e q u i p m e n t 17,086 t o n s
CoMo c a t a l y s t 2 kg Weight of p u m p s a n d 2.565 t o n s
Air 1 3 7 8 ~ 103Nm3 c o m p r e s s o r s
Water 1384 X l o 3 m3 Weight of pipes 9,671 t o n s
REFERENCES
Bellman, R. (1961) Adaptive Control Processes
-
a Guided Tour. Princeton University Press, Princeton, New Jers.ey. Report CHEM/SEM.B/R.l6. Chemic+l Industry Committee of the United Nations.Dobrowolski, G., Kopytowski, J., and Zebrowski, M. (1980a) S y s t e m s Analysis in Management of Optimal S t r a t e g y of Growth. Paper presented at the 3rd IFAC/IFORS Conference, 16-19 June, Warsaw. In preprints of the 3rd IFAC/IFORS Conference on Dynamic Modelling and Control of National Economies, published by the Polish Academy of Sciences. Copies available from the Institute of Sys- t e m s Research (Polish Academy pf Sciences), Warsaw.
Dobrowolski, G., Kopytowski, J., and Zebrowski, M. (1980b) Development forecast of production structure. In the Proceedings of t h e XN Union of Polish Architects Congress. 9-14 June. Zakopane. Poland. pp. 19-26.
Dobrowolski, G., Kopytowski, J.. Lewandowski, A., arid Zebrowski. M. (1982) Generating EjFcient Alternatives f o r Development of t h e Chemical h d w t r y . Collaborative Paper CP-82-54, International Institute for Applied Systems Analysis, Laxenburg, Austria.
Gallagher, J.M., and Zimmerman, R.G.J. (1978) Coal in l'ransition: A n Analysis of k v e l o p m e n t Programs Using t h e Bechtel h h e r g y S u p p l y P l a n n i n g Model. Pub- lished by Research and Engineering Bechtel National Inc., San Francisco, Cali- fornia. Prepared for the Energy Modeling Forum (ENP), 3. February 1978.
Gorecki. H. (1981) Problem of choice of an optirnal solution in a niulticriterion space.
In the Proceedings of t h e VIIIth W i e n n i a l World IFAC Congress, 24-28 August.
Kyoto, Japan, pp. 106-110.
Gorecki, H., and Kopytowski, J. (1980) Polyoptinral Approach to t h e Problems of Design of t h e P r o d u c t i o n S t r u c t u r e in t h e Chemical h z d w t n j . Paper presented at the 5th Italo-Polish Seminar on the Application of Systems Theory t o Economics, Management and Technology, Torun. Copies.available from t h e authors.
Gorecki, H., Dobrowolski, G., Kopytowski, J., and Zebrowski. M. (1982) The quest for a concordance between technologies and resources as a multiobjective decision process. In M. Grauer, k Lewandowski, and k P . Wierzbicki (Eds.), k l t i o b j e c t i v e a n d Stochastic Optimization. Collaborative Proceedings CP-82-S12. lnterna- tional Institute for Applied Systems Analysis, Laxenburg, Austria.
Grenon, M.. and Lapillone, B. (1976) m e WELMM Approach t o h r g y Strategies and Options. Research Report RR-78-19, lnternatiorlal Institute for Applied Systems Analysis. Laxenburg, Austria.
HZfele, W.. Barnett. H.. and Sassin, W. (1982) Kiinftige fossile Brennstoffe
-
lhre Nutzung und Einbettung in moderne Energiesysteme (F'uture Fossil Fuels-
Their Utilization a n d Embedding in Modern Energy Systems).
DFYLR-Nachrichten. Mitteilungsblatt der Deutschen Forschungs- und Versuch- sanstalt fiir Luft- und Raumfahrt E.V., 35: 8-19.
Kallio. M., Lewandowski. A., and Orchard-Hays. W. (1980) An h p l e m e n t a t i o n of t h e B f e r e n c e Point Approach f o r Multinbjective Optimization. Working Paper WP- 80-35, International Institute for Applied Systerr~s Analysis. Laxenburg. Austria.
Kendrick, D.A., and Stoutjestijk, A.J. (1978) The R a n n i n g of h d u s t r i a l h v e s t m e n t P r o g r a m s . Vol. 1 . A Methodology. Published by Johns Hopkins University Press, Baltimore, M.D.. for World Bank Research Publications.
Kopytowski, J.. Wojtania, J.. and Zebrowski, M. (1981) Fossils as K e y R e s o u r c e s of H y d r o c a r b o n s f o r t h e Chemical h d u s t r y
-
t h e B u r n i n g P r o b l e m of h d u s t r i a l W v e l o p m e n t . Collaborative Paper CP-81-20, International Institute for Applied Systems Analysis, Laxenburg. Austria.Nowak, H. (1982) P r o s p e c t s f o r Methanol P r o d u c t i o n . Unpublished report (in Polish).
Institute for Industrial Chemistry. Cracow. P o l a n d Based on C h e m i c a l Market
& p o r t , Vol. 215, No. 266, p.5 (1979); C h e m i c a l Age, Vol. 118, No. 3122, p. 10 (1979); Sherwin (1981) - s e e below; h f o r n . C h e m i e . No. 166, p. 181 (1977).
Rogner. H.H. (1982) S u b s t i t u t i o n of Coal a n d Gos f o r Oil - S o m e Global Considera- t i o n s . Paper presented a t t h e first US-China Conference on Energy. Environ- m e n t and Resources, 7-12 November. Beijing. Copies available from the Energy Group a t t h e lnternational Institute for Applied Systems Analysis. Laxenburg.
Austria.
Sassin. W. (1982) Fossil h h e r g y a n d i t s A l t e r n a t i v e s - a P r o b l e m B e y o n d Costs a n d Prices. Paper presented a t t h e lnternational Economic Association, Conference on Economics and Alternative Sources of Energy, 27 September-1 October, Tokyo. Japan. Copies available from t h e Energy Group a t the lnternational Institute for Applied Systems Analysis. Laxenburg, Austria.
Sherwin, Martin B. (1981) Chemicals from methanol. H y d r o c a r b o n Processing 60(3):
79-84.
Sophos. A., Rotstein, E., and ~ t ~ ~ h a n o ~ o l o u s , G. (1980) Multiobjective analysis in modeling the petrochemical industry. C h e m i c a l E n g i n e e r i n g S c i e n c e 35(12):
24 15-2426.
Stadtherr, M.A., a n d Rudd, D.F. (1976) Systems study of t h e petrochemical industry.
C h e m i c a l E n g i n e e r i n g S c i e n c e 31(11): 1019-1028.
St. Pierre. L.E. (Editor) (1978) R e s o u r c e s of Organic M a t t e r f o r t h e f i t u r e . Perspec- t i v e s a n d R e c o m m e n d a t i o n s . Multiscience Publications Limited, Montreal.
Canada.
Wierzbicki, AP. (1980) ?'he Use of R e f e r e n c e O b j e c t i v e s in Midtiobjective O p t i m i z a t i o n
-
Theoretical I m p l i c a t i o n s a n d Practical b p e r i e n c e . Working Paper WP-79-66.International lnstitute lor Applied Systems Analysis, Laxenburg, Austria.