In situations where discontinuity in operation occurs, specifically in a country where coontractualization has an increasing trend, the performance level of operators after the work break is of great interest. Existin...In situations where discontinuity in operation occurs, specifically in a country where coontractualization has an increasing trend, the performance level of operators after the work break is of great interest. Existing studies have found that the performance of an operator declines after her operation is completely stopped. However, when the operator performed other tasks （may it be similar or not from her previous task） during the work break, the performance after the work break seems to be affected at different level. Contractual and regular operators from a semiconductor and textile company were considered to replicate a discontinuous and continuous operation. The processing times of contractual workers before and after several months of work break were compared. Two types of work break were seen to have significant effect on an operator＇s performance after the work break, Type 1： 0% to 40% similarity from previous task and Type 2： 40% to 97% similarity from previous task. One can find that when 21% of tasks performed during the work break are similar to the operator＇s previous task, there would be no change in her performance upon returning. On the other hand, a 5% decline in performance was observed after work break type 1 and an 8.54% improvement after work break type 2. Also, a remission rate of 18% from end of stint 1 to start of stint 2 under work break type 1 was seen, while 8% for work break type 2. This may also be true to other industries. Thus, further study is suggested.展开更多
In operations management, the learning curve has been an effective tool in estimating operator performance. However, discontinuities in work disrupt the learning process and a phenomenon called remission or forgetting...In operations management, the learning curve has been an effective tool in estimating operator performance. However, discontinuities in work disrupt the learning process and a phenomenon called remission or forgetting occurs, resulting in increased time of performing the task upon resumption of work. The study aims to identify variables that significantly affect the remission rate of sewing operations in a manufacturing setting. Four variables--length of stint 1, percent Differenceat stint t, gender, and product family, were identified. Statistical analyses, such as paired t-test, correlation, regression, and analysis of variance （ANOVA） were conducted in order to observe the relationships between the dependent variable and independent variables. For the results of the first general regression, gender was found to be an insignificant variable in predicting remission rate, while product family, length of stint 1, and percent Differenceat stint I were statistically significant. Moreover, the final general regression, which excluded the insignificant gender variable and considered the （regrouped） product families, revealed that product family, length of stint 1, and percent Differenceat stint 1 were still statistically significant. Length of stint 1 had a moderately positive correlation with remission rate, while percent Differenceat stint i had a moderately negative correlation with remission rate. Also, percent Differenceat stint 1 was the largest contributor to the remission rate model. In terms of R2, the goodness-of-fit of the model is moderate. Finally, the model yielded an absolute error of 5.08%, indicating a high accuracy in predicting remission rate.展开更多
文摘In situations where discontinuity in operation occurs, specifically in a country where coontractualization has an increasing trend, the performance level of operators after the work break is of great interest. Existing studies have found that the performance of an operator declines after her operation is completely stopped. However, when the operator performed other tasks （may it be similar or not from her previous task） during the work break, the performance after the work break seems to be affected at different level. Contractual and regular operators from a semiconductor and textile company were considered to replicate a discontinuous and continuous operation. The processing times of contractual workers before and after several months of work break were compared. Two types of work break were seen to have significant effect on an operator＇s performance after the work break, Type 1： 0% to 40% similarity from previous task and Type 2： 40% to 97% similarity from previous task. One can find that when 21% of tasks performed during the work break are similar to the operator＇s previous task, there would be no change in her performance upon returning. On the other hand, a 5% decline in performance was observed after work break type 1 and an 8.54% improvement after work break type 2. Also, a remission rate of 18% from end of stint 1 to start of stint 2 under work break type 1 was seen, while 8% for work break type 2. This may also be true to other industries. Thus, further study is suggested.
文摘In operations management, the learning curve has been an effective tool in estimating operator performance. However, discontinuities in work disrupt the learning process and a phenomenon called remission or forgetting occurs, resulting in increased time of performing the task upon resumption of work. The study aims to identify variables that significantly affect the remission rate of sewing operations in a manufacturing setting. Four variables--length of stint 1, percent Differenceat stint t, gender, and product family, were identified. Statistical analyses, such as paired t-test, correlation, regression, and analysis of variance （ANOVA） were conducted in order to observe the relationships between the dependent variable and independent variables. For the results of the first general regression, gender was found to be an insignificant variable in predicting remission rate, while product family, length of stint 1, and percent Differenceat stint I were statistically significant. Moreover, the final general regression, which excluded the insignificant gender variable and considered the （regrouped） product families, revealed that product family, length of stint 1, and percent Differenceat stint 1 were still statistically significant. Length of stint 1 had a moderately positive correlation with remission rate, while percent Differenceat stint i had a moderately negative correlation with remission rate. Also, percent Differenceat stint 1 was the largest contributor to the remission rate model. In terms of R2, the goodness-of-fit of the model is moderate. Finally, the model yielded an absolute error of 5.08%, indicating a high accuracy in predicting remission rate.
