Determining a combination of factors on rubber end manufacturing using taguchi method at PT. Agronesia Bandung

Rubber end is one of products that manufactured by PT. Agronesia Bandung. These products experienced defect and unconformity to the standard which is bubble trapped in rubber end products. This defect can be found from 3 factors – control factor, which are material thickness, temperature, and shift. Material thickness has 3 levels, which are 3 mm, 5.5 mm, and 7 mm; temperature has 3 levels, which are, 35° C - 40° C, 40° C - 45° C, and 45° C - 55° C; while shift has 3 levels, which are shift 1, shift 2, and shift 3. To solve the problem, Taguchi method is used to have the right combination of those 3 factors to produce rubber end. First step is selecting factors and determining level that will be a base for selecting orthogonal matrix, and then determining S/N ratio and analysis of variance. The result is factors combinations are material thickness at level 2 of 5.5 mm, temperature at level 3 of 45° C - 55° C, and shift at level 1 of shift I.


Introduction
PT. Agronesia is a rubber manufacture, and one of their product is rubber end. The company experience the problems on this rubber end that are defects on the products. These defects are bubble trapped, bare, raw, exceed dimension, and thickness, and these defects causing the products will be reworked or even rejected. Figure 1 shows that the most occurred defect is bubble which is 8 occurrences from 16 or 50%, so that this defect needs a special attention to reduce defects ratio and to improve the product quality.

Method
In business, engineering and manufacturing, quality has a pragmatic interpretation as the noninferiority or superiority of something; it's also defined as fitness for purpose. Quality is a perceptual, conditional, and somewhat subjective attribute and may be understood differently by different people. Consumers may focus on the specification quality of a product/service, or how it compares to competitors in the marketplace. Producers might measure the conformance quality, or degree to which the product/service was produced correctly [2].

Experimental design
Experimental design is the design of any task that aims to describe or explain the variation of information under conditions that are hypothesized to reflect the variation. The term is generally associated with true experiments in which the design introduces conditions that directly affect the variation, but may also refer to the design of quasi-experiments, in which natural conditions that influence the variation are selected for observation [4].
In its simplest form, an experiment aims at predicting the outcome by introducing a change of the preconditions, which is reflected in a variable called the predictor (independent). The change in the predictor is generally hypothesized to result in a change in the second variable, hence called the outcome (dependent) variable. Experimental design involves not only the selection of suitable predictors and outcomes, but planning the delivery of the experiment under statistically optimal conditions given the constraints of available resources.

Taguchi method
The Taguchi method of quality control is an approach to engineering that emphasizes the roles of research and development, product design and product development in reducing the occurrence of defects and failures in products. The Taguchi method considers design to be more important than the manufacturing process in quality control and tries to eliminate variances in production before they can occur [8].
Taguchi refers to experimental design as "off-line quality control" because it is a method of ensuring good performance in the design stage of products or processes. Some experimental designs, however, such as when used in evolutionary operation, can be used on-line while the process is running [7].

Factors, factors level 2.3.1 Factors
Based on defects pareto, it can be determined that the independent variable is bubble trapped due to the most occurrence defect. For dependent variable (factors) can be determined from the early researches will be material thickness, temperature, and shift.

Factors level Material thickness
The material thickness is significantly influence for bubble trapped in rubber end. The product thickness is 20 -21 mm, so that material thickness will be set in 3 levels as follows, • Level 1 = 3 mm • Level 2 = 5.5 mm • Level 3 = 7 mm

Temperature
There are 3 levels for this factor, as follows Shift is the interval work to produce rubber end. There are 3 levels for this factor, as follows By applying Taguchi method, the quality of manufactured goods, and engineering designs are developed by studying variations. The Composite Principle Component has been optimized by using Taguchi method. Analysis of variance (ANOVA) has been conducted for Composite Principle Component to find the optimal process parameters. Signal to Noise (S/N) Ratio has been found for PCA to find the optimal levels of the process parameters. Finally, a conformation test has been made for three different materials and the results have been plotted. [3] The Taguchi method was used to determine optimum conditions for tire rubber in asphalt concrete with Marshall Test. The tire rubber in asphalt concrete was explored under different experimental parameters including tire rubber gradation (sieve #10-40), mixing temperature (155-175 1C), aggregate gradation (grad. 1-3), tire rubber ratio (0-10% by weight of asphalt), binder ratio (4-7% by weight of asphalt), compaction temperature (110-135 1C), and mixing time (5-30 min). The optimum conditions were obtained for tire rubber gradation (sieve #40), mixing temperature (155 1C), aggregate gradation (grad. 1), tire rubber ratio (10%), binder ratio (5.5%), compaction temperature (135 1C), mixing time (15 min). [5]

Result and Discussion
Before orthogonal matrix is constructed, first the degree of freedom (DoE) has to be determined. DoE is meant to calculate the number of minimum experiments to inquire the observed factors.  In this experiment, orthogonal matrix L9 (3 4 ) is chosen, so that a linear graphic standard which is necessary to the experiments as follows, From the linear graphic in figure 2., it can be seen that factor A will be placed in column 1, factor B will be placed in column 2, factor C will be placed in column 3, and Error will be placed in column 4. The number of 1, 2, 3 are randomize process. Level on the orthogonal matrix column was made in number 1, 2, 3, etc. Or it can be done by using own method to describe the level, for example 0 and X, (+) and (), L and H, to replace the number of 1, 2, 3 [6].
To find the factors that influence to bubble trapped, it needs to analyze by signal to noise (S/N) ratio in the analysis of variance. When the mean of responds was found, then the next step is to determine a robust design which is a design that will not affected by noise factor. The S/N which is been used is a quality characteristic, the smaller the better [6]. The quality of bubble trapped is good when the smallest bubble trapped is the better. The mean of factor effect based on S/N ratio is an average of S/N respond from each particular levels and factors, the calculation was done based on the respond of experiments.  B3 C1 Table 5 shows that the setting of effected factors is factor C level 1, factor B level 3, and factor A level 2.
As the influence factors to the mean of bubble trapped were found, then to find a significant influence factors to S/N ratio can be done by construct a two-way analysis of variance model.