It is known that potholes are generated on pavement surfaces when snow thaws or it rains rather than when the weather is fine. If a pothole causes an accident or the like, road management will be deemed to be defective. In this study, the authors propose the rules for intensifying management according to conditions so that the pothole management level is increased according to the weather condition. In detail, the authors express the process of pothole generation, which depends on the rainfall condition, with a Poisson generation model, and the transition mechanism of pothole generation process due to the weather with a Markov switching Poisson process model. The authors also propose a model estimation method using the Markov chain Monte Carlo method considering the latent variable sampling with the multi-move sampler. Lastly, the effectiveness of the proposed rules for intensifying pothole management for expressways is discussed.

View full abstract

 The fluctuation of the price of construction materials has an influence on the viability of construction projects, and is one of the key risks to manage. In the case of PFI (private finance initiative) projects, the risk is more serious because the period between contract signing and the procurement of materials is longer than that with conventional projects. The purpose of the present study is to establish a quantitative risk analysis system with which more rational risk sharing regarding construction materials can be investigated between the public and the private sectors. Structural equations determining the prices based on related economic variables are built for seven major construction materials with the data recorded by month for 15 years. With additional time-series analysis of the economic variables, a Monte Carlo simulation system is built to estimate the total cost of construction projects. By applying the system to some model PFI projects, it is found that the present prevailing rule of risk sharing in Japan is biased in favor of the public sector.

View full abstract

 With the purpose of streamlining and quantifying maintenance and management, the long-term monitoring with sensors is being put into practice. On the other hand, methodologies for detecting abnormalities and extracting the progress of deterioration based on monitoring data have not been developed. In this study, the authors propose a method for predicting deterioration based on the statistical characteristics of time-series data obtained through long-term monitoring. In detail, the fluctuation in the variance of the error term of the time-series model is expressed by the ARMAX-GARCH regression model, which is developed by the ARMA model with trend. Then, a methodology of estimating the timing of detailed inspection utilizing the ARMAX-GARCH model is developed. Lastly, this methodology is applied to the data of long-term monitoring targeted at the joint members of viaducts, to evaluate its effectiveness.

View full abstract

 This study analyzes tendering data on bid and wins prices for general civil engineering work by Rank A and B in 2002-2011 based on information obtained from Shikoku Regional Development Bureau, Ministry of Land Infrastructure, Transport and Tourism. Regression analysis by using factors such as the predetermined price and number of participants reveals that a major part of bid prices is explained by the predetermined price and that if the number of participants increases, then the decrease in both bid and win prices is statistically significant. Further, businesses' behavior seems to have changed structurally since early 2006, and if the regression includes a structural changing effect (competition dummy), then the coefficients of bid and win prices are negative but not statistically significant. Therefore, on including the competition factor, the comprehensive evaluation system is likely to decrease price by itself. However, the coefficient of the bidding ratio is observed to be four times larger than that of the technological factor in the comprehensive evaluation system based on bidding probability analysis.

View full abstract

 The production system of Toyota, one of the biggest companies in Japan, has been studied and introduced in many manufacturing companies in Japan and overseas because of its high productivity. Additionally, in foreign countries, this production system has been applied to construction works, as so-called “Lean Construction”, at many construction sites. However, this method almost never has come into use in Japan.
 For the purpose of better understanding of Lean Construction in Japan, the authors have proposed a new definition of Lean Construction on the base of the current Toyota Production System (TPS). Further, exploring the possibility of implementing Lean Construction, the authors have investigated the examples which were implemented in Japan and were considered to use the Lean Construction method without being aware that they are Lean Construction. The authors also have made a study on the types of works and site conditions that are fit for implementation of Lean Construction.

View full abstract