Journal of Applied Research on Industrial Engineering
Volume:1 Issue: 4, Autumn 2014

Data Envelopment Analysis (DEA) is used for calculating of relative efficiency and then ranking of decision making units (DMU) subject to inputs and outputs in a continuous decision making space. In this paper, DEA models, Cook and Kress Model and Belton-Vickers model have been applied as a multi attribute decision making (MADM) tool and compared analytically to other MADM models such as simple additive weighted (SAW) and technique for order preference by similarity to ideal solution (TOPSIS). For this purpose, after simulating some decision making matrixes and replacing DMU with alternatives, outputs with criteria to be maximized, inputs with criteria to be minimized, alternatives will be ranked by these models. The results of this study show that incorporating decision maker value judgments into the DEA models (restricted DEA models) provide comparable results to traditional MADM models (such as SAW and TOPSIS that are applied more than others). So, restricted DEA models seem to be an advantageous tool for solving MADM problems.

The attestation towards environmental, statutory obligations and economic interests arising from rehabilitation operations in recent years has led to more focus on reverse logistics operations. To this end, integration of the design of reverse and forward logistics networks which results in prevention of sub optimality due to separated design of these networks is of high significance. This model discusses a mixed-integer nonlinear programming model for the integrated design of multi-level and multi-commodity forward-reverse supply chain network. In the end, the calculation results of the proposed model solution have been presented via GAMS software in order to locate facilities, determine the relationship between facilities and raw material procurement rate and production rate. The result of the solved multi-purpose models has corroborated the single-purpose models and it shows the efficiency of the used methods.

Organization's competitiveness and efficiency depend on the reliability, availability and productivity of their production equipment today. In order t o increase reliability and safety level with reasonable cost, it is vital for companies to select the appropriate maintenance strategy. Among all the duties of maintenance departments, having a proper maintenance strategy developed based on the needs of the organization and aligned with the organization's goals is a necessity. It is also considered the first and most important step toward the department's success. This paper presents a hybrid approach of analytical network process and DEMATEL to select optimal maintenance strategy. Results revealed that the best maintenance strategy for transformer equipments of Exir Pharmaceutical Company of Boroujerd is reliability centred maintenance.

In this paper, the problem of scheduling in a two-stage supply chain system is investigated with considering coordinated scheduling of different stages. The objective is to find the similar sequences of processing tasks in two steps with two aims of minimizing the total preparation time as well as the maximum preparation time of each stage. To do so, the problem has been solved by non-dominant sorting genetic algorithm (NSGA-II) and using a variety of multi-objective techniques. The proposed meta-heuristic algorithm is applied for the first time in this problem. The results show that the proposed algorithm is able to find Pareto optimal solutions with a high efficiency.

Timetabling is one of the most difficult issues in the world; this is a combinatory optimization issue and it has been proven that it is a NP-Hard issue.
University Courses Timetabling is very important especially for the exams and courses. The manual solving of the Timetabling needs a broad domain of sources and time to create an applicable schedule with minimum interference in the curriculum and Professor's program is not easy. Different mathematical models and algorithms have been presented for this issue but each strategy considers different limitations according to its environment and factors. Timetabling is different for different university courses Timetabling. In this study a new multi objective mathematical model with new objective functions and new constrains has been presented for university courses timetabling. The present study tries to consider most restrictions for a training center. Finally the validity of the proposed model has been surveyed by a small numerical example and its solving by LINGO11 software shows that this model can satisfy all goals and limitations.