Practical Adaptive Method for Heat Sealing Parameter Selection Using a Response Surface Methodology Approach

Abstract

Response surface methodology was followed to construct a predictive model for the strength of film to gland seals in liquid liner bags bonded through an established heat sealing process. The method provides a structured alternative to the current trial and error approach. Once the response surface has been trained, the manufacturer can determine the effect of process changes to a known level of confidence. The model was further used to investigate how much the known process parameters contribute to variance of the established process. A model constructed using sealing head temperature, actuation pressure and dwell time is able to account for 90%of the total observed variance. Independent validation tests were conducted to verify the model, results obtained were within the constructed 95%confidence interval, and the mean approximated the true response. The model accounted for 89% of the observed variance of the validation set and is suitable for a generalized prediction.