CBR and Machine Learning for Combustion System
Design

Jutta Stehr
Daimler Benz AG
Dept. F3S/E
Postfach 2360
89013 Ulm, Germany
email: stehr@dbag.ulm.DaimlerBenz.com



Abstract
Nowadays the automotive industry has to face two major challenges. First products must meet
continually increasing government requirements on fuel economy and low exhaust emission.
Second the market demands product variety and short production cycles. The automobiles
combustion system determines the exhaust emission rate, combustion system engineering is
one of the crucial steps in the development process. Cylinder head design is a good example of
showing how enhanced AI technologies like CBR and Machine Learning support high-level
eingineering design tasks.

The work described was coordinated in a joint project between the Daimler-Benz research
group on Thermo and Fluid Dynamics and our reasearch group on Machine Learning with the
aim of improving of cylinder head engineering. This paper proposes how Machine Learning
and specifically Case-based Reasoning (CBR) transform a traditional database containing both
geometry and air-motion data into a so called experience memory for cylinder head design. We
will present the initial steps of our database analysis in terms of different learning algorithms,
then use the extracted knowledge to develop case-based design retrieval and quality prediction
modules.
5. References

[ACO92]	T.Acorn, S.Walden. SMART: Support Management cultivated
reasoning technology for Compaq customer service. Proceedings of
AAAI-92. MIT Press. 1992.

[AHA91]	D. Aha, D. Kibler, M. Albert. Instance-Based Learning Algorithms.
Machine Learning 6 (1). 1991.

[AHA94]	D. Aha, R. Bankert. Feature Selection for Case-Based Classification of
Cloud Types: An empirical comparison. Workshop Notes on Case-
Based Reasoning. AAAI-94. 1994.

[BAR89]	R. Barletta, D. Hennessy. Case Adaptation in Autoclave Layout Design.
DARPA Case-Based Reasoning Workshop. Morgan Kaufmann
Publishers. 1989.

[BOE9O]	C. D. de Boer, R. J. R. Johns, D. W. Grigg. B. M. Train. I. Denbratt.
J.R. Linna. Refinement with perforamnce and economy for four-valve
automotive engines. Society of Automotive Engineers. 1990.

[BR089]	D. C. Brown, B. Chandrasekaran. Design Problem Solving: Knowledge
Structures and Control Strategies. Pitman. 1989.

[CAR93]	C. Cardie. Using Decision Trees to Improve Case-Based Learning.
Proc. of the 10th Int. Conference on Machine Learning. Morgan
Kaufman. 1993.

[CHE94] Cheshire Engineering Corporation. Neuralyst. Users Guide. 1994.

[C0S93]	S. Cost, S. Salzberg. A weighted Nearest Neighbor Algorithm for
Learning with Symbolic Features. Machine Learning 10 (1). 1993.

[D0R76]	D. Drner. Problemlsung als Informationsverarbeitung. Verlag W.
Kohihammer. 1976.

[FAB92]	FABEL-Consortium. Survey of FABEL, Fabel Report No. 2, GMD,
1992.

[HAM89]	K. Hammond. Chef: A Model of Case-Based Planning. Proceedings of
the Nat. Conference on Artificial Intelligence (AAAI 86). MIT Press
1986.

[KEL91]	J. D. Kelly, L. Davis. A Hybrid Genetic Algorithm for Classification.
Proc. of the 1 2zh Int. Joint Conference on Artificial Intellicence. 1991.

[KOL93]	J. Kolodner. Case-Based Reasoning. Morgan Kaufmann Publishers.
1993.

[LIN77] P. H. Lindsay, D. A. Norman. Human Information Processing.
Academic Press. 1977.

[MIC94] D. Michie, D. J. Spiegeihalter, C. C. Taylor (eds.). Machine Learning,
Neural and Statistical Classification. Ellis Horwood. 1994.

[PEA92]	M. Pearce, A. K. Goel, J. L. Kolodner, C. Zimring, L. Sentosa, R.
Billington. Case-Based Design Support. IEEE Expert. October 1992.

[PIA91]	G. Piatetsky-Shapiro, W. J. Frawley (eds.). Knowledge Discovery in
Databases. MIT Press. 1991.

[POL49] G. Polya. Schule des Denkens. Francke Verlag. 1949.

[SIM92]	E. Simoudis. Using Case-Based Retrieval for Customer Technical
Support. IEEE Expert. October 1992.

[QUI93]	J. R. Quinlan. C4.5 Programs For Machine Learning. Morgan
Kaufmann Publishers. 1993.

[TAK87]	H. Takahashi, T. Ishida, K. Sato. Improvement of Diesel Emgine
Performance by Variable Swirl System. Int. Off-Highway &
Powerplant Congress. Society of Automotive Engineers. 1987.

[WAL91]	J. P. Walsh, G. R. Ungson. Organizational Memory. Academy of
Management Review. 16 (1). 1991.

[WES91]	S. WeB. PATDEX/2 - ein System zum adaptiven, fallfokussierenden
Lernen in technischen Diagnosesituasitionen. SEKI Working Paper
SWP-91-01. Universitt Kaiserslautern. 1991.

[WIR95]	T. Reinartz, R. Wirth. The need for a Task Model for KDD. Accepted
by the MLnet Familiarisation Workshop on Machine Learning,
Statistics and Knowledge Discovery in Databases. 1995.
