Journal of Operational Research in its Applications
Year:2016 | Volume:13 | Issue:1 (48)|Papers Count:10

In supply chain management, the supplier's performance will be evaluated by several criteria. In this paper, a fuzzy multi-objective mathematical programming is designed to select the best suppliers as well order allocation with considering qualitative and quantitative factors as well as risk. At the first step, potential suppliers were evaluated by AHP algorithm and multi-objective modeling has been done in a fuzzy manner considering a number of factors. The purposed model was solved by MOPSO algorithm, and achieved answers from MOPSO were ranked by TOPSIS algorithm. Finally, parameters sensitivity analysis of the model has been done.

Governmental agencies are recurrently trying to reallocate the additional resources of the oil industry and taxes to provinces/ industrial sectors usually in a macro socio-economic environment, and at the second level the agent banks allocate credit to loan applicants (companies) in a micro socio-economic environment considering the upper levels allocations. The upper level has its own objectives and constraints including sustainable development and a minimum country GDP growth and etc. The lower levels also have their own objectives and constraints including minimizing their credit risk while maximizing their profit. In order to handle the issue, a stackelberg game is designed which uses a mathematical portfolio programming model in its first level allocating credit to different industry sectors and provinces. At the second level a credit scoring algorithm is used for building the agent banks model in order to allocate credit to loan applicants (companies). Fuzzy apriority is used as the classification algorithm in order to classify the loan applicants to good and bad ones. The classifier results are the initial answer of the lower level mathematical problem that should meet the objectives and constraints of the lower level, which is described in terms of a liner multi-objective programming. Using genetic algorithm as a reliable metaheuristic algorithm is inevitable for changing the classifiers parameters and producing different initial answers to fit the agent banks' needs. A combination of fuzzy apriority and traditional genetic algorithm bi-level based algorithm (GABBA) is designed and introduced here that is named Genetic algorithm fuzzy apriority bi-level based algorithm (GAFABBA) which is used to solve the problem. The validation of the results is tested using three other different credit allocation scenarios.

Nowadays, method of decision-making for selection of a project in an organization is very important, especially with being aware of this point that most of the organizations are involved in this issue, and even sometimes the biggest part of their income is earned from such projects. By using AHP, this research aims at identification and prioritization of effective criteria of selection of investment projects in the product department of a company which is active in steel industry. In this research, effective criteria on selection and evaluation of investment projects were identified through observation, studying, and interviewing experts, and then effective criteria of investment projects were weighed by the AHP method. Required information have been collected through the distribution of questionnaires and pair comparisons among company specialists and experts and through interviewing them. For evaluating projects, two groups of financial-economic and technical criteria were identified. The weights of criteria and sub-criteria were obtained depending on authorities' ideas and from their pair comparisons. With regard to the acquired results, the main financial-economic criteria were more important than the technical ones.

DEA is a technique that measures the efficiency of decision-making units (DMUs) based on their use of inputs and outputs. Believing that future planning and predicting the efficiency of DMUs is very important, this paper first presents a new multi-objective fuzzy stochastic DEA model (MOFS-DEA) with common weights in dynamic environment under mean chance constraints that considers the fluctuations of data per repeated periods. In the initial proposed MOFS-DEA model, the inputs and outputs are assumed to be characterized by random triangular fuzzy variables with normal distribution in which data are changing sequentially. Under this assumption, the solution process is very complex, so we then convert initial proposed MOFS-DEA model to its equivalent multi-objective stochastic programming. In order to shorten the computational time, we convert the equivalent multi-objective stochastic programming model to mono-objective stochastic model using the fuzzy multiple objectives programming approach. To solve it, we designed a new hybrid algorithm by integrating Monte Carlo (MC) simulation and genetic algorithm (GA). Finally, one case study is presented to demonstrate the proposed modeling idea and the effectiveness of the hybrid algorithm. The computational results show that our hybrid algorithm outperforms the hybrid GA algorithm which was proposed by Qin and Liu (2010) in terms of runtime and solution quality.

In the design of gas network pipes, several conditions must be considered such as the environmental conditions, special pipe conditions, initialization cost, etc. The optimal gas network is designed considering these running conditions. Running conditions are the design and placement of various systems such as pressure reduction stations, routing of pipelines and determining their diameter based on the locations of consumers; here, the optimal distribution network is designed. In this paper, a model is presented to design an optimal gas network such that investment costs are minimized. The proposed network contains city gate stations, town distributed stations and consumers. Here, the location of town-distributed stations, routing and diameter determining of pipes are modeled simultaneously. Some of the constraints are nonlinear which are approximated by linear constraints. The proposed model is solved by Cplex12.5 for a real case study, and results are reported.

Data envelopment analysis (DEA) is a technique for the evaluation of decision-making units (DMUs) with multiple inputs and outputs. In DEA, each DMU selects favorable weights to obtain the most efficient score. But the nature of self-evaluation and flexibility in DEA to choose the optimal weight of DMUs has been criticized a lot; therefore, cross-efficiency methods have been introduced in DEA to overcome such criticisms. In addition, DMUs produce desired outputs along with undesired ones such as greenhouse gases, and for this models have been presented to evaluate DMUs with undesirable outputs. The aim of this paper is to present methods for the evaluation of the cross-efficiency of DMUs with undesirable outputs.Therefore, with the extension of existing models in the undesirable outputs literature, several secondary goal models are presented for evaluation of cross-efficiency of DMUs with undesirable outputs. The presented models are based on multi-objective programming. Finally, it has been tried to illustrate the importance of the issue with a real-world example.

This research calculates the efficiency and productivity indexes of high-speed railways and after determining them compares the status of efficiency and productivity of countries that have access to such high-speed railways. The data envelopment analysis (DEA) and multi objective decision-making process are used for calculating the efficiency and productivity, respectively.The existing literature and statistical data are used as input data for determining the key parameters of the models. This research explores various approaches of DEA efficiency calculation, that is, the BCC and CCR methods. Each method is implemented by identifying all input-output criteria and by considering the input-oriented prospective for evaluating the efficiency of high-speed railways in various countries. Applying the method of Anderson-Peterson, the efficiency of the four efficient countries (which obtained the unit value of efficiency) are further examined and the final ranking of the countries is determined. Afterward, a multi-criteria decision-making process is established for productivity cannulations. In order to build the hierarchical process, three basic criteria, that is, technical specifications, operational performance, and economic parameters are selected as key indicators/ criteria. Each of these criteria is divided into three sub-criteria, and the multi-criteria decision-making analysis has been implemented on the entire system. Eventually, the ranking of various countries with respect to the efficiency and productivity of high-speed railways is determined. The results of analysis show that Japan, China, France, and Spain-in order mentioned here- have obtained both the best efficiency and productivity of the high-speed railway system.

In recent decades, production planning has been considered in many studies as one of the most important practices for achieving organizational objectives. Recently, Data Envelopment Analysis (DEA) has been applied as an effective approach to evaluate the relative efficiency of Decision Making Units (DMUs) and for developing production plans in terms of resource allocation and target setting. This paper reviewed the existing literature on DEA-based production planning and developed a network production planning model which takes the magnitude size of DMUs into account. Since the production information cannot be precisely measured in some cases, the uncertain theory has played an important role in production planning. Thus, the main contribution of this paper is to develop a fuzzy network DEA-based approach for target setting in a centralized decision making environment, when the next production demands are uncertain. The new proposed model takes into account desirable and undesirable outputs simultaneously. To illustrate the applicability of the proposed model, a data set of 13 production units in Guilan provinces is used. The results indicate that the proposed model is potentially beneficial.