We propose a shared taxi dispatching method that utilizes demand prediction information. In this method, a taxi picks up passengers if the delay of the arrival time of other passengers that use the same taxi is within the defined period. Each taxi selects a parking lot to stop by using demand prediction information and information about other taxis' parking lot occupation so that passengers can be picked up quickly. The proposed method was evaluated by simulation and reduced passengers' waiting time compared to a method in which each taxi selects a nearest parking lot.
We propose a user interface for inducing communication of regional electronic coupon system. The proposed interface will be introduced at an event held for the purpose of revitalizing the shopping street. Three types of user interface candidates are designed. The best one is selected through preliminary user and shop staff survey. The selected user interface named “With Touch” were experimentally used an actual event, “BunjiBar” in Kokubunji city, Tokyo, in 2019. Good users' comments are given such as “it was fun to make a payment with a shop staff”. The result shows that the proposed interface is effective for inducing communication of regional electronic coupon system.
In this paper, we propose an efficient method of technical structure analysis of FPGA development using network analysis. Recently, with the reform of the industrial structure called “Industry 4.0” and the explosive expansion of the application area of AI (Artificial Intelligence), the demand for computing power in edge devices is increasing. For these backgrounds, the acceleration by FPGA is focused on the industry. Hence, in this paper, we analyze the technology development trends of a significant FPGA supplier. As analytical data, we employ patents obtained by the company in the U.S. We focus on the classifiers in the patent information. We create graphs by classifiers and analyze their clusters. From the results of extracting graph features by the proposed method, key technology areas and their transitions are revealed in the technology development of FPGA suppliers.
The price of crops depends on the grade. The grade is determined from various factors such as color, shape, size, and scratches. It takes a lot of time and effort if farmers decide the grade. This study aims to automate shape grade judgement. In the prior method, binarization and contour extraction are performed on the tomato image to define features called flatness, variant, sharpness, and irregularity, and a threshold is set for each grade. However, due to the correlation between the features, the grade judgement has not been made with sufficient accuracy. In this study, we propose several methods focusing on contour features and compare these accuracies.
Monte Carlo tree search is used in AlphaGo Zero which is a strong artificial intelligence player for Go, and therefore it has attracted much attention. It can be used for searching for the optimum of combinatorial optimization problems, and it is promising for finding the optimum or a near-optimum. This paper proposes a Monte Carlo tree search method for the knapsack problem which is one of the typical combinatorial optimization problems. In the proposed method, a new candidate solution is generated by using superior ones found so far in the procedure called the simulation. Its performance is evaluated through conducting numerical experiments.
MES (Manufacturing Execution System) is often used in pharmaceutical industry to make manufacturing process sure. And, it brings higher quality products. On the other hand, it is difficult for pharmaceutical companies to identify its benefit quantitatively due to several reasons such as various conditions or over estimation of benefits. This paper shows an efficient approach to estimate cost benefit of implementing MES by integrating qualitative analysis with quantitative analysis. The proposed approach was successfully applied to an Indian pharmaceutical company's case.
Discussion how to reach a root failure point in open systems problem. In most case, the analyzing process depends on the experiences of support engineers, but there is no known way for trouble shooting in such circumstance. The function chain method enables to draw down a problem visually to show the direction of analyzing process. It also provides efficient manner to drill down the problem to detail of failure point. Secondary we propose the method how to judge whether the components on the function chain have been failed or safe. It stands on to compare the log information that captured pre and post of the problem. Finally, we show the efficiency of function chain by analyzing the actual open systems problem reports. The result proofs that there is a superiority of the function chain with comparison method rather than Fault Tree Analysis. Also, it shows the function chain method reduces the number of analyzed components. There is a possibility that one function chain has multiple failure points. However, the majority of function chains have a single failure point. Therefore, in this study, we only discuss the single failure problem.
In this paper, we explore the ideal way of urban parks for a sustainable society. Urban park has many functions that make comfortable space in urban. Expanding urban areas and increasing population in urban bring some problems. Urban parks are decreasing these problems. So urban parks are important places of comfortable urban life. However, some issues are there. Small budget, difficult to understand the actual use of urban parks, and Urban parks are aging so on. We use text mining to analyze intention information on the internet — analyzed intention information to help understanding make better urban parks. Estimated intentions and behavior provide important suggestions for efficient urban park maintenance.