| Popis: |
Rubber material has been used in footwear sole manufacturer for many years due to its low price and provides better physical properties. Vulcanization process is one of rubber sole manufacturing processes that affects the quality of rubber sole. Elongation at break (E) and slip resistance (COF) are some responses that are used to evaluate the performance of vulcanization process. The quality characteristics of these responses are “higher-is-better”. The optimization was conducted by using backpropagation neural network method and particle swarm optimization method. Three important process parameters such as mold temperature, mold pressure, and holding time were used as input parameters. Each process parameter was set at three different levels. Hence, a 3 x 3 x 3 full factorial was used as design experiments, the experiments were replicated three times along with randomizations. The architecture of developed BPNN network had 3 neurons, which are input layer, 1 hidden layer with 12 neurons and 4 neurons in output layer. The activation functions of hidden layer, output layer and network training were tansig, purelin, and trainlm respectively. PSO optimization showed the optimal conditions were 220.93% for elongation at break and 0.174 for slip resistance, at these optimum points of mold temperature, the mold pressure and holding time were 161°C, 87 bar and 4 minutes, respectively. The total reducing loss cost of the process was IDR 507/unit or 30% of loss cost before optimization. |
