Production time estimation in task center refraction for detail production areas with machine learning algorithms

In today's competitive conditions, it is very important for companies that want to have a competitive advantage to use limited resources efficiently, to estimate production time to identify future investments. But traditional methods, such as time study that comes to mind when it comes to production time, require a large workload in facilities where the range of products and processes is too high, resulting in a loss of time and therefore cost. At this point, production time estimation with machine learning algorithms provide a great advantage in terms of time and cost. In this study, it is aimed to estimate production times in task center breakdown in detail production areas of a production facility with machine learning algorithms. The use of artificial neural networks, support vector regression and gradient boosting machine algorithms, which have been successful results as prediction methods, has been decided. In the rest of the study, the algorithm that gave the most successful results for each task center was determined based on the average absolute percentage error value from performance metrics. According to the results obtained, the artificial neural networks gave an error rate of 28,22% on average, while the support vector regression gave an error rate of 15,92% and the gradient boosting machine gave an error rate of 14,69%.