Methods to identify, quantify and minimise variation of net weights in canned foods : a thesis presented in partial fulfillment of the requirements for the degree of Master of Technology, Faculty of Engineering and Technology, Massey University, New Zealand

Abstract

Using a 24 factorial design model, the methods to identify and quantify the major sources of variation of net weights in canned foods were investigated. A piston filler was selected using modified starch solution as the filling medium. The stroke length and speed of the filler and the concentration and temperature of the filling medium comprised the four factors. The data were transformed into means and variances of fill weights from both across filling heads and across consecutive filling cycles, and were used as responses. The responses, which were derived across filling cycles for each of the filling heads, were used as blocks, to evaluate the head effects. The projection designs were used to optimise variation and fill levels at set piston-stroke levels. The factor level combinations required to minimise variation and maximise fill level which was computed through a model matrix using all important effects were found to be P+,S+,T-,C+ and P+,S+,T+C+ respectively. The contributions of factors and their interactions to the short-term variance of fill weights were estimated using variance across heads within consecutive filling cycles (88.5%). The analysis across filling cycles within individual heads estimated the deviation of fix factor levels within the trials and contributed to 44%, which appeared as factor effects. Most of this variation (52.3%) which was caused by the unstable filling mechanisms appeared as the residual error. The analysis of blocks using heads was successful in partitioning the variance due to head differences (3.6%). The high volume operations generated a higher contribution from unstable filling mechanisms to the total variance, and a lesser contribution from head differences to the total range of fill weights. The recommendations include methods and materials to reduce the error in the design. Future research is recommended in the areas of vacuum and single shot fillers, multifilling processes, and particle size variation.