Presentation of Risks Analysis Model in Urban Projects Based on Data Mining Technique with Case Study

Analysis of the right response to risk is one of the important processes in project management. The purpose of this research is to categorize the risks of the urban projects. To this end, after identifying the risks of the urban project, the most important indicators are developed in line with experts’ opinions to evaluate risks. These include impact on time, cost, quality, probability of occurrence, environmental impact, safety effects, importance of risk, risk manageability and risk response strategy. Then, the risk assessment is performed using the desired indicators. All steps are implemented according to CRISP-DM standard methodology and the importance of risk, risk manageability, and risk response strategy are predicted by data mining algorithms. The results show that classification algorithms performed in risk management successfully. Importance of risk and risk manageability are predicted by logistic regression whose accuracy rates are respectively equal 0.88 and 0.9. For risk response strategy, the Naïve Bayes algorithm performed better than other algorithms with an accuracy rate of 0.84. For further investigation of the used algorithms, the results are compared with one of the MCDM methods, the TOPSIS method. Data mining algorithms performed better than the TOPSIS method.