A data compressed ART-1 neural network algorithm [group technology application]

Summary form only given, as follows. Adaptive resonance theory (ART) neural networks are being developed for application to the industrial engineering problem of group technology. Two and three dimensional representations of engineering designs are input to ART-1 networks to produce groups or families of similar parts. These representations, in their basic form, amount to bit maps of the part, and can become very large when the part is represented in high resolution. An enhancement to the algorithmic form of ART-1 to allow it to operate directly on compressed input representations and to generate compressed memory templates has been developed. The performance of this compressed algorithm was compared to that of the regular algorithm on real engineering designs. Significant savings in memory storage as well as a speed-up in execution were observed