Application Issues of Genetic Programming in Industry

Kordon, Arthur K.; Castillo, Flor; Smits, Guido; Kotanchek, Mark

doi:10.1007/0-387-28111-8_16

Cited by 23 publications

(12 citation statements)

References 4 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…An example of this approach with many successful applications in the chemical industry is by using symbolic regression generated by Pareto Front Genetic Programming [33]. Most of the implemented empirical models are very simple (see examples in [34] ) and were easily accepted by process engineers. Their maintenance efforts were low and the performance over changing process conditions was acceptable.…”

Section: G Emerging Simplicitymentioning

confidence: 99%

Future Trends in Soft Computing Industrial Applications

Kordon

2006

2006 IEEE International Conference on Fuzzy Systems

View full text Add to dashboard Cite

The paper summarizes the current state of the art of applying soft computing solutions in the chemical industry, based on the experience of The Dow Chemical Company and projects the future trends in the field, based on the expected future industrial needs. Several examples of successful industrial applications of different soft computing techniques are given: automated operating discipline, based on fuzzy logic; empirical emulators of fundamental models, based on neural networks; accelerated new product development, based on genetic programming, and advanced process optimization, based on swarm intelligence. The impact of these applications is a significant improvement in process operation and in much faster modeling of new products.The expected future needs of industry are defined as: predictive marketing, accelerated new products diffusion, highthroughput modeling, manufacturing at economic optimum, predictive optimal supply-chain, intelligent security, emerging simplicity, and handling the curse of decentralization. The future trends that are expected from the soft computing technologies, which may satisfy these needs, are as follows: perception-based modeling, integrated systems, universal role of intelligent agents, models/rules with evolved structure, and swarm intelligence-based process control.

show abstract

Section: G Emerging Simplicitymentioning

confidence: 99%

Future Trends in Soft Computing Industrial Applications

Kordon

2006

2006 IEEE International Conference on Fuzzy Systems

View full text Add to dashboard Cite

show abstract

“…Several examples of real world symbolic regression problems include chemical process modeling [8] [9], approximation and rainfall runoff modeling [2], data mining [6], design of experiments [1], and economic decision analysis [10] [4].…”

Section: Brief Background Of Symbolic Regressionmentioning

confidence: 99%

Dynamics of evolutionary robustness

Piszcz

Soule

2006

Proceedings of the 8th Annual Conference on Genetic and Evolutionary Computation

View full text Add to dashboard Cite

Recently there has been considerable interest in determining whether, and how much, evolutionary pressure for genetic robustness influences evolutionary processes. In this paper, we attempt to show that this evolutionary pressure does have a significant effect in typical genetic programming problems. Specifically we demonstrate that in a standard genetic programming implementation to solve a symbolic regression problem, pressure for genetic robustness forces the population away from high fitness, but less robust, solutions in favor of solutions with lower fitness, but higher genetic robustness.

show abstract

“…However, the growing interest of industry in GP [5] requires a more systematic approach for the GP model generation process to guarantee consistency of delivered results. An important part of this process is the appropriate parameters setting for each specific type of applications, which will improve the efficiency of model development and minimize the development cost.…”

Section: Introductionmentioning

confidence: 99%

“…The best way to address this issue is by using statistical Design Of Experiments (DOE) [6] on industrial data. Fortunately, as a result of the current successful GP applications, a set of industrial benchmark data sets has been collected (a summary for the applications in the chemical industry is given in [5]). They are with different sizes and data quality and each one is Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page.…”

Section: Introductionmentioning

confidence: 99%

Pareto front genetic programming parameter selection based on design of experiments and industrial data

Castillo

Kordon

Smits

et al. 2006

Proceedings of the 8th Annual Conference on Genetic and Evolutionary Computation

Self Cite

View full text Add to dashboard Cite

Symbolic regression based on Pareto Front GP is the key approach for generating high-performance parsimonious empirical models acceptable for industrial applications. The paper addresses the issue of finding the optimal parameter settings of Pareto Front GP which direct the simulated evolution toward simple models with acceptable prediction error. A generic methodology based on statistical design of experiments is proposed. It includes statistical determination of the number of replicates by half-width confidence intervals, determination of the significant inputs by fractional factorial design of experiments, approaching the optimum by steepest ascent/descent, and local exploration around the optimum by Box Behnken or by central composite design of experiments. The results from implementing the proposed methodology to a small-sized industrial data set show that the statistically significant factors for symbolic regression, based on Pareto Front GP, are the number of cascades, the number of generations, and the population size. A second order regression model with high R 2 of 0.97 includes the three parameters and their optimal values have been defined. The optimal parameter settings were validated with a separate small sized industrial data set. The optimal settings are recommended for symbolic regression applications using data sets with up to 5 inputs and up to 50 data points.

show abstract

Application Issues of Genetic Programming in Industry

Cited by 23 publications

References 4 publications

Future Trends in Soft Computing Industrial Applications

Future Trends in Soft Computing Industrial Applications

Dynamics of evolutionary robustness

Pareto front genetic programming parameter selection based on design of experiments and industrial data

Contact Info

Product

Resources

About