In recent years, the movement to introduce robots in production systems with the aim of replacing manual labor or combining manual and robot labor has started. In particular, dual-arm robots, which have two robotic arms, are gaining much attention in view of their ability to replace manual laborers. However, planning an efficient work system beforehand and teaching the lean movements to the robots are essential for using them more effectively. In assembly processes, the result of selecting the assembly order greatly influences the productivity of the production process.
In such a background, this paper examines a method for determining an assembly order with high work efficiency. Under the assumption that dual-arm robots assemble the products that can be assembled easily (i.e., stacking toy blocks is used here), we propose a computational model for searching a highly efficient assembly order utilizing reinforcement learning. We also consider a method for using the results of previous learning model studies to effectively find solutions for new assembly models. Regression analysis is utilized to transfer the past learning results.
The quality loss proposed in studies by Taguchi (1985, 1986) has been recognized as a new evaluation measure of quality instead of using the proportion of nonconforming items known as the traditional evaluation measure of quality. In recent years, various kinds of acceptance sampling inspection plans for assuring lot quality under the concept of the quality loss have been developed. In this paper, we propose a variable two-stage sampling inspection plan based on the quality loss as a new type of double sampling inspection plan that is terminated at finite sampling inspection from the viewpoint of practical usefulness. At the same time, reducing the average sample number required for sampling inspection is considered in the newly developed variable two-stage sampling inspection plan. The design procedure for the developed sampling inspection plan is established, and then usefulness of the sampling inspection plan is shown through numerical comparison.
Traditional demand forecasting methods are categorized as scientific methods (e.g., time series analysis or regression analysis) or methods based on experience and tacit knowledge (e.g., delphi method or market research). Recently, research that combines these forecasting methods has become a hot topic and categorized as a demand forecasting method based on the prediction market. It is known that the prediction market was able to accurately forecast the vote ratio for the US presidential election. In the field of supply chain management, the research is applied to forecast the future demand of products. In this study, we propose a demand forecasting method that uses a voting system based on a collective intelligence mechanism. We examine forecasting accuracy using real business data from a five-month period. According to statistical tests, we show that the forecasting method we propose performs more accurately than the existing method used in our company.
In the printer manufacturing industry, it is known that business impact is especially large because of the due date penalty or opportunity loss (including sales loss of toner cartridge and drum cartridge) when an inventory shortage occurs. Furthermore, in this industry, due to occurrence of sudden and large demand, it is not clear how to analyze demand distribution from historical sales data and how much inventory should be kept. We reviewed various literature regarding the demand forecasting method for sudden and large demand, but we were unable to find a proper method, and therefore decided to develop a new method ourselves. In this research, we propose a new demand forecasting method based on extreme value distribution. In the method proposed, we first classify historical sales data, then we estimate the parameters of the extreme value distribution, and finally forecast future demand. Additionally, we tested our method using real business data. As a result, we found that the method proposed performs better than the current method being used by businesses and can avoid inventory shortages when sudden and large demand occurs.