Prediction of Highway Construction Delay due to Weather Impact Using Fuzzy-Probabilistic Risk Analysis

Highway construction projects are conducted in an outdoor environment, and are therefore sensitive to various weather conditions, especially rainfall. Rainfall is regarded as a major risk factor, which often leads to partial or complete suspension of construction project operations. In this paper, a decision support system is developed for quantifying the impact of rainfall and temperature on highway construction schedules. The system incorporates a database that uses historical daily rainfall and temperature data alongside a knowledgebase that includes a fuzzy inference system, employing fuzzy rules acquired from experts in highway construction. As the monthly delays are random variables, probability theory is also applied for frequency analysis. This system can be useful in preparing realistic schedules for the preconstruction stage and in analyzing weather-related claims for the post construction stage. It can also be used in predicting the impact of short term weather forecasts in the real- time construction process. To demonstrate the usefulness of the system, it has been applied to the cities of Tehran and Gorgan in Iran, which have very different climates, and the results are compared and discussed. The results show how the differences in climate of these cities and the time and stage of construction are reflected in predicted construction delays.