utility function

Various techniques and approaches have been presented to solve multi-objective decision making problems under different assumptions. The STEM technique is also one of the most widely used techniques for dealing with this group of problems, especially for solving multi-objective linear programming (MOLP) models. In the current research, a new approach has been proposed as the development of the mentioned method. In this regard, the second phase of the method is integrated with the simulation process and the concept of utility function has been used in order to determine the probability of selecting targets. For each of the selected targets, several random adjustment rates are defined. Random selection of satisfactory functions based on their utility and using simulation tools, will "create diversity in the selected objectives for adjustment". Also, by considering different and random adjustment rates for selected functions, while facilitating the decision-making process and providing an analytical report to the decision-maker, it will be possible to apply "different levels of satisfaction" of the decision-maker. A three-objective problem with five constraints, is solved using the proposed method and its results are presented. The results indicate that applying the above changes will lead to solving some of the basic limitations of this method that have been mentioned in previous studies. Comparing the proposed technique with the basic method, shows the overall superiority of the proposed method, especially in the criteria related to interaction with the decision maker.

Considering the large volume of mineral resources in Iran and the geographical extent of mines and related industries in the country, its transportation is of particular importance. Although this product is known as a rail-friendly product in terms of transportation mode, but the 39% share of the rail transportation industry is far from the capabilities and expectations of this field. Therefore, in order to increase the share of rail transportation, it is necessary to identify the factors in the attractiveness of rail transport compared to road one. Identifying the factors in mode choice of transportation is one of the most important issues that has always been of interest to researchers in this field. Several studies have been conducted on understanding the behavior of customers to choose the mode of transport. In this paper, the most important factors that affect the mode choice have been identified and the demand function for minerals rail transport is estimated.

For this purpose, at first step, the affective factors in choosing the mode of transportation have been identified with the revealed preference method through literature review and background studies. Using the multiple regression model and the SPSS software, the correlation between the dependent variable (share or demand) and independent variables have been measured and a list of variables that effect on attracting demand and have a significant presence has been identified. In the end, using the double logit model and NLOGIT software, the rail utility function has been estimated to absorb the demand for minerals. The logit model is a discrete choice model, which in this research includes a utility function with different choice options and parameters affecting them. The structure of these models is of the probability type and in it the behavior of the decision maker and his efforts to maximize the utility resulting from the choice are modeled through mathematical relationships. The "forward selection" method has been used to build the model. In this method, the independent variables are entered into the model one by one in the order of their influence on the dependent variable and are evaluated, and at each stage, according to their t test statistic sign, chi-square test statistic and model fit. The next step is to obtain the final model of the utility function. Finally, the validity of the model was measured by the in-sample method, which shows a 3% error in demand forecasting

The results show that the most important factors affecting the choice of mineral transportation method are the ratio of rail transportation tariff to road tariff per ton, the bulk of the mine load, and the accessibility of the origin and destination to the rail network. This means that with the increase in the price of the rail tariff compared to the road, the desirability of rail transportation decreases and the probability of transferring cargo to the road increases. Also, having origin and destination access to the rail network was identified as one of the favorable factors for the rail transportation. This can be included in network development policies for investing in connecting major cargo centers to the network. On the other hand, the lack of access to the network increases the cost of combined transportation, which, in addition to wasting time, causes additional time and unloading and loading operations, which will lead to a decrease in the attractiveness of rail. Load accumulation with a threshold of 200 million ton-km increases the possibility of attracting demand to the rail side. Also, by increasing the ratio of rail to road distance on a fixed route, the possibility of rail and road competition decreases.

The results of this model help the transportation service providers to focus on improving the identified factors that lead to the growth of their performance in the transportation market of this product.

The application of neural networks in estimating and describing the structure of decision makers' priorities has been very much considered in solving multi-objective decision problems in recent years. The neural network is a new approach to estimate the decision-making function of a multi-objective problem. Developing and improving the teaching methods of these types of networks facilitate to find the preferred solution in multi-dimensional issues, especially large-scale issues. In this paper, the educational method is developed to increase the efficiency of a neural network. In addition, a genetic algorithm is used to train this neural network. Furthermore, an improved cost function is proposed to adjust the parameters of the neural network and based on this function the cost parameters of the neural network are optimized. The efficiency of the proposed method is shown in solving several practical examples, including linear/nonlinear and discrete/continuous optimization problems. The efficiency of the proposed method is shown in solving several practical examples, including linear/nonlinear and discrete/continuous optimization problems.

Iran has experienced vast variation of exchange rate and its destructive effects during The last twenty years. Also, recently most of investors, importers and banks have sustained losses because of extreme increase of Euro price and fluctuations of other exchanges. Also, it seems that Iranian investors, beside the return criterion do not consider risk criterion so much, or that they do not pay enough attention to it as an important criterion for investment. So in this paper, we present a set of interobjectives of risk and return trade-offs along with an analysis of Iranian Sepah Bank investment in an allocated exchange portfolio and use of credit approach of Weighted Global Criterion (WGC) with assumption P = 2, ∞ to optimize the model of bi-criterion and we will present that this method is better than utility function method of Markowitz. Also, the results explain that bi-criterion model is consistent with Iranian exchange investment policy (based on fewer concentrations on USA Dollar).