SYSTEMS FOR THE PREDICTION OF PROCESS PARAMETERS

Manufacturers of machine tools, cutting tools and electrodes assist us ers by providing some form of performance data with their products. Ho wever, production engineers do not have resources and time to study ma nuals or performance charts and do detailed analysis to predict the pe rformance of the system for the chosen operating parameters. Thus, an attempt has been made to develop a performance-prediction system for o perating parameters such as machinability data selection, which will a ssist process planners and machinists in their decision-making process es. A back-propagation neural-network model is proposed for prediction . The use of traditional empirical models and expert systems for machi nability studies is also discussed. The proposed method will also help in interpolating/extrapolating collected factual data from the shop-f loor, for different machining conditions.