---

Knowing customer behavior patterns, clustering and assigning them is one of the most important purpose for banks. In this research, the five criteria of each customer, including Recency, Frequency, Monetary, Loan and Deferred, were extracted from the bank database during one year, and then clustered using the customer's K-Means algorithm. Then, the multi-objective model of bank service allocation was designed for each of the clusters. The purpose of the designed model was to increase customer satisfaction, reduce costs, and reduce the risk of allocating services. Given the fact that the problem does not have an optimal solution, and each client feature has a probability distribution function, simulation was used to solve it. In order to determine the neighbor optimal solution of the Simulated Anneling algorithm, neighboring solutions were used and a simulation model was implemented. The results showed a significant improvement over the current situation. In this research, we used Weka and R-Studio software for data mining and Arena for simulation for optimization. The results of this research were used to develop Business Intelligence software for customers in one of the private banks of Iran.

---

---

Every project has many risks and as there are many complexities in projects today, recognizing the most important risks is essential for projects' success and efficiency. In this research, we tried to determine most significant risk's categories in the framework of risk breakdown structure of 4th edition of Project Management Body of Knowledge Guide that can be generalize to all projects in Iran. With considering dependencies and interactive relations between risks of project, we used DEMATEL method to determine the most significant project risk's categories on the basis of risk breakdown structure of 4th edition of Project Management Body of Knowledge Guide. Also fuzzy set theory was applied to measure experts' subjective judgments, experts who have rich expertise and knowledge in Iranian projects were selected to evaluate the influences. The results revealed that "External", "Technical", "Project Management" and "Organizational" risks are significant and in the most important risk's category which is "External", "Regulatory" risks and in "Technical", "Project Management" and "Organizational" risks, "Technology", "Estimating" and "Project Dependencies" are the most important risks respectively and should be paid more attention because they were in the first rank of importance.

---

Mathematical models have the potential to provide a cost-effective, objective, and flexible approach to assessing management decisions, particularly when these decisions are strategic alternatives. In some instances, mathematical model is the only means available for evaluating and testing alternatives.  However, in order for this potential to be realized, models must be valid for the application and must provide results that are credible and reliable. The process of ensuring validity, credibility, and reliability typically consists of three elements: verification, validation, and calibration. Model verification, validation and calibration are essential tasks for the development of the models that can be used to make predictions with quantified confidence. Quantifying the confidence and predictive accuracy of model provides the decision-maker with the information necessary for making high-consequence decisions. There appears to be little uniformity in the definition of each of these three process elements. There also appears to be a lack of consensus among model developers and model users, regarding the actions required to carry out each process element and the division of responsibilities between the two groups. This paper attempts to provide mathematical model developers and users with a framework for verification, validation and calibration of these models. Furthermore, each process element is clearly defined as is the role of model developers and model users. In view of the increasingly important role that models play in the evaluation of alternatives, and in view of the significant levels of effort required to conduct these evaluations, it is important that a systematic procedure for the verification, validation and calibration of mathematical models be clearly defined and understood by both model developers and model users.