High-end pack ages used for AI applications are becoming denser with the emergence of chiplets technologies. Among them, the density of package substrates is also progressing, and the recent substrates tend to have an asymmetrical substrate structure. However, such an asymmetrical substrate induces warping due to heating process of chip joining, and it is essential to control the copper remaining ratio in the substrate in order to suppress it at the design stage. In this paper, a genetic algorithm is used to optimize the copper remaining ratio, and an algorithm flow considering the allowable warpage value at chip bonding is proposed. As a result of the actual optimization evaluation, the superiority of the proposed flow was confirmed.

View full abstract

 For improving product safety and reliability, possible defects are analyzed in advance using FMEA, FTA, and so on. However, whether such methods can predict defects depends on the knowledge and experience of engineers. In this paper, we propose a defect cause analysis method that can automatically derive possible causes without relying on the experience of engineers, and can extract causes difficult to predict. First, events and their causal relationships are modeled with Petri nets. In addition, an assembly model is introduced to represent the propagation of events between parts. Next, a method to predict the causes of product defects using the causal relationship model and the assembly model is proposed. In the method, a model including not only the functions of a product but also its manufacturing process is created at first. Then, past event causal data related to events in the product model are combined into the model. After that, a defect cause prediction model could be generated by reversing the relationship between cause and effect. By using this model, a cause event of a product defect can be traced step by step. Thus, possible underlying causes of the product defect are automatically derived. Finally, we confirmed the usefulness of our proposed method by implementing a system and predicting the causes of a defect of a soldering iron.

View full abstract

 Recent advances in computer technology and information and communication technology are realizing highly-distributed manufacturing systems (HDMSs) where each machine is computerized and can communicate with other machines. For this type of manufacturing system, a distributed method of discrete event simulation and that of production scheduling were proposed, respectively. Because these methods have the same architecture based on common distributed sequencing algorithms for HDMSs, they can be integrated and the integration makes it possible to generate a better operation plan efficiently. Considering this point, an integrated method of the distributed scheduling and simulation for HDMSs was developed. This paper discusses enhancement of this method. In the integrated method, several dispatching rules for job selection and machine selection are utilized and the weights of the rules are optimized using simulated annealing by an external monitoring system in a centralized manner. This optimization process is improved so that it is performed in a decentralized manner by each machine with synchronization by the monitoring system. The optimization process is enhanced further so that not only non-delay schedules but active schedules can be searched. The effectiveness of this enhanced method was shown by numerical experiments.

View full abstract

 As systems become more complex, the evaluation criteria for optimizing them increase. This paper aims to develop a multi-objective hierarchical optimization method for solving optimization problems of large-scale systems. The proposed multi-objective hierarchical optimization method requires a method for evaluating the Pareto solution of the upper hierarchy using the Pareto optimal solution of the lower hierarchy. In this paper, we propose an index to improve the quality of the Pareto solution based on the number of optimal solutions that update the existing Pareto solution in the Pareto front of the upper hierarchy, which is created by integrating the Pareto solutions of the lower hierarchy. Furthermore, we evaluate the maximum or minimum value of the upper-level Pareto front as an indicator for improving the spread of Pareto solutions. The results show that using a numerical model, the proposed method achieves a wider Pareto front than the conventional method in a case study.

View full abstract

 To study the relationship between struct ure and thermal conductivity of thermal conductive grease, it is essential to evaluate the structure under pressure, which is the actual use environment. However, evaluating the cross-sectional shape under pressure is difficult as grease is in a semi-fluid state. In this study, a novel method of preparing and observing a cross-section of thermal conductive grease under pressure has been developed using a focused ion beam-scanning electron microscope. An aluminum foil/ grease/copper plate was pressurized with a vise, and the sample including the vise was immersed in liquid nitrogen to freeze and fix the structure. Results indicated that the thickness of grease and the distribution of thermal conductive particles in the grease could be quantitatively evaluated by this method. We found that the packing ratio of thermal conductive parti cles in the grease under pressure remained consistently high, similar to the level before pressurization.

View full abstract