After the Great East Japan Earthquake, industrial companies that store a certain amount of stocks, i.e., their finished products and unfinished products, are increasing. It is because they want to continue their business and to fulfill responsibility to supply the products to the market. Based on above-mentioned circumstances, practical method and criteria for estimating suitable amount of stocks are needed. This paper discusses evaluation method of recovery process from earthquake damage for manufacturing line consists of plural manufacturing apparatuses (components). And its points are; considering the finite amount stocks that are stored between each manufacturing process, and taking impacts of the damage correlation between components into account. In addition, using the proposed method, recovery curves for an example of virtual manufacturing line including the stocks are estimated, and the damage correlation influence is clarified through the difference among the curves shape.

View full abstract

In this study, Adachi-ku Senju district in Tokyo was selected as the investigated district because Senju district has been designated by the Tokyo Metropolitan Government as the area with both high fire and flood risk. Questionnaire survey was conducted for the residents, employees and students in Senju district to collect the data regarding the knowledge of damage estimates of Senju district, evacuation method and evacuation behavior at the disaster occurrence. As the results of analysis, it became clear that there are several problems in Senju distrct such that the employees have a tendency not to accurately understand the damage estimates and the evacuation method, both the residents and the employees have a tendency to underestimate the flood risk, and some residents try to evacuate from the flood by going up the lower floor of the residence under the situation that fire spread and flood are occurring at the same time.

View full abstract

To evaluate the effects of velocity inhomogeneity in propagation path on seismic ground motion, numerical simulations were performed using inhomogeneous medium models with fluctuation in seismic velocity. In this study inhomogeneity effects in short distance region from seismic source are focused on. Epicentral distance is about less than 50km and is approximately equal to 50 times ratio to the analyzed central wave length. 3D inhomogeneous models were generated and a double-couple type point source was located in the medium models. Then the finite difference method was applied to the models to obtain velocity records at observation points on the surface. The simulated records were revealed to have coda waves similar to observed records and to have modulated wave forms and amplitudes differed from calculated velocity records in a homogeneous medium model. Such effects were observed even in relatively short distance. Fluctuation in amplitudes was observed in simulated waves in inhomogeneous medium even if the source radiation effect was removed. The fluctuation was revealed to depend on the inhomogeneous parameters and frequency. The ratios of the amplitudes in inhomogeneous medium to those in homogeneous medium were decreasing with increasing distance to show the scattering attenuation. The effect of source radiation pattern on the scattering attenuation has not been seen. Also the fluctuation in amplitudes is seen to increase gradually with distance. As a result even in relatively short distance, such as several tens of kilometer, the effects of inhomogeneity in velocity can cause effects on wave forms and amplitude fluctuation.

View full abstract

Conventionally, in wide-area prediction of strong ground motion or long-period ground motion by the finite difference method (FDM), it is difficult to evaluate the ground motion considering a top surface low-velocity layer with S-wave velocity as low as 100m/s from the limitations of the computer memory for calculation. We attempted to devise a modeling method for minimizing the amount of calculation in the FDM, and investigated a simplified numerical analysis method that can be considered from the seismic bedrock to the ground surface. To be specific, a so-called "Equivalent shallow ground model" obtained by averaging the ground physical properties of several layers shallower than the engineering bedrock was provided in a virtual manner, and the ground motion was calculated by the FDM using an “Integrated ground model” from the seismic bedrock to the ground surface. Furthermore, the ground motion obtained from the integrated ground model was pulled back to the engineering bedrock by using the equivalent shallow ground model based on the multiple reflection theory. Finally, the ground motion at the ground surface was recalculated from the engineering bedrock using the original (true) shallow ground model. In addition to this modeling method, in order to reduce further the calculation amount in the FDM, a so-called "Uniform shallow ground model" that reflects directly the ground physical property of engineering bedrock to the surface layer was provided virtually, and a method for calculating the ground motion in a manner similar to that shown above was also examined. As a result, in both examined methods, as compared with the conventional so-called "Separation analysis" separating the analysis region at the engineering bedrock, it has become possible to further evaluate the precise ground motion. By performing an “Integrated analysis” from the seismic bedrock to the ground surface, it is possible to improve the accuracy of long-period ground motion in addition to the short-period ground motion.

View full abstract

In order provide basic data for strong ground motion distribution of the recent damaging earthquakes, the occurrence areas of destructive ground motion are collected from isoseismal maps of the earthquakes and their characteristics are examined. The results show that the areas of seismic intensity more than 6 lower almost agree with the previous earthquake data while the areas of seismic intensity 7 are smaller. We also confirmed that the distance from the earthquake fault and the local ground condition largely affect the occurrence of the destructive ground motion.

View full abstract

We investigated tombstone fall-down rates from November 30 to December 2 in 2014 to estimate the distribution of the ground motion intensity around the near source region of the 2014 Northern Nagano Prefecture Earthquake (M_J6.7), Japan. In addition, we conducted microtremor observations and evaluated horizontal-to-vertical spectral ratios (H/V) to interpret the spatial distribution of the intensity in terms of ground vibration characteristics. In Hakuba-mura Horinouchi where wooden houses were heavily damaged, tombstone fall-down rates were 100%. This result corresponded to seismic intensity 7 (in the 0-7 Japanese scale) according to previous studies. H/V amplitudes of this area were found to be significantly different from other areas such as Hakuba-mura Oide or Otari-mura, and this difference can be considered to be due to the difference in subsurface geology. In Hakuba-mura Oide where obvious surface faults were found, the tombstone fall-down rates of hanging-wall side were 13-56% (averaged seismic intensity 6-), and those of foot-wall side were 0-23% (averaged seismic intensity 5+). We also investigated tombstone fall-down rates also in Otari-mura, Ogawa-mura and Nagano-shi. Based on these results, we showed the spacial variation of the ground motion, corresponding to seismic intensity from 5- up to 7, around the near source region.

View full abstract

A probabilistic fault displacement hazard analysis (PFDHA) is a methodology that assesses the annual rate/probability of exceedance that an amount of displacement of a surface earthquake fault exceeds a certain quantity. In Japan, Takao et al. (2013) and Takao et al. (2014) basically proposed evaluation formulae for PFDHA based on investigation results related to earthquake faults that had occurred in Japan, model experiment results and numerical simulation results, but some of the coefficients followed Youngs et al. (2003). In this paper, some probability density function parameters for the distributed fault were revised on the basis of the maximum likelihood estimation method.

View full abstract

[in Japanese]

View full abstract

[in Japanese]

View full abstract