Predicting extrusion process parameters in Nigeria cable manufacturing industry using artificial neural network.

Area of Science:

• Materials Science
• Manufacturing Engineering
• Computational Intelligence

Background:

• The thermoplastic extrusion process involves numerous complex and fluctuating parameters.
• Current methods for determining extrusion parameters rely on expensive and time-consuming trial-and-error techniques.
• There is a need for more efficient and accurate methods to optimize extrusion processes in manufacturing.

Purpose of the Study:

• To determine realistic extrusion process parameters in Nigeria's cable manufacturing industry using artificial neural networks (ANNs).
• To enhance the efficiency of thermoplastic extrusion operations by predicting process parameters before plant execution.
• To bridge the gap between theoretical analysis and real-world manufacturing systems.

Main Methods:

• Development of an artificial neural network model in a MATLAB environment.
• Training the ANN using a supervised learning method, specifically the Levenberg-Marquardt Algorithm.
• Utilizing real manufacturers' data from Nigerian cable industries for model training and validation.

Main Results:

• The developed ANN model demonstrated capability in predicting manufacturing process parameters.
• The model can accurately forecast parameters for various grades of PVC thermoplastic material.
• The ANN approach offers a significant improvement over conventional trial-and-error methods.

Conclusions:

• Artificial neural networks provide an efficient and effective tool for optimizing thermoplastic extrusion parameters.
• The study successfully demonstrated the practical application of ANNs in Nigerian cable manufacturing.
• Predictive modeling using ANNs can lead to cost savings and reduced lead times in industrial extrusion processes.