نشریه بین المللی مهندسی صنایع و مدیریت تولید
سال بیست و پنجم شماره 1 (بهار 1393)

In project scheduling, the activity durations can often be reduced by dedicating additional resources. Time/Cost Trade-off Problem considers the compromise between the total cost and project duration. The discrete version of the problem assumes a number of time/cost pairs, so called modes, and selects a mode for each activity. In this paper, we consider the Discrete Time/Cost Trade-off Problem. We study the problem of minimizing total cost subject to a deadline on project duration so called Deadline problem. Projects are often subject to various sources of uncertainties that have a negative impact on activity durations and costs. Therefore, it is crucial to develop effective approaches to generate robust project schedules that are less vulnerable to disruptions caused by non-controllable factors. We propose a novel robust model for the DTCTP in which interval uncertainty is assumed for the unknown cost parameters. We use genetic algorithms to solve the proposed robust optimization model. Our computational results on large-sized problem instances have revealed the satisfactory behavior of our algorithms.

In today's conditions that many domestic banks of country are experiencing a kind of credit risk, establishing a credit risk management system is necessary to reduce bank's outstanding claims and resolve the failure to repay loans problem of central bank. Creating a proper costumer's credit ranking system is one of the fundamental issues and requirements of establishing a credit risk management system. Five C's of credit system is on of the reliable systems for costumers ranking which can be used for this purpose. On the other hand, effort to achieve a suitable tool for implementing and executing this system is an important and inevitable issue. VIKOR method is one of the capable multi-attribute decision making methods which can be applied for executing five C's of credit system and solving the costumer's credit ranking problem. In this paper, a VIKOR method is presented which not only is capable of determining the optimum values of importance weights of criteria but also could take into consideration fuzzy importance weights of decision maker's judgments during the process of costumer's credit ranking. The proposed method is adopted to solve a numerical example about credit ranking of bank's costumers and thereby, the best alternative for giving loan facilities is selected in uncertain conditions.

In this paper, a general procedure is developed to find all non-dominated solutions of the multi objective combinatorial optimization (MOCO) problem. This procedure is based on the elastic constraints method and applies the identification of objective's bounds for it. The bounds of objectives are determined by solving single objective integer programming problems. First, the elastic constraints method is performed on a bi-objective and tri-objective problem respectively, then it is developed on a general multi objective integer programming (MOIP) problem. In this paper, a numerical example such as tri-objective assignment problem is presented to clear the proposed method.

In the steel industry that involves with large amount of wealth, labor and energy flow a small improvement in planning has significant consequences such as reducing costs, energy consumption, production time, delivery time and also increasing customer satisfaction. This research aim to present a efficient model in order to schedule the orders in the product line of cold rolled in the Mobarakeh Steel Company. The model is solved by applying real data of Mobarakeh Steel cold rolling in two stages. At first the third model without considering zero-one constraint is solved. Then the zero-one and some other constraints obtained from heuristic algorithm are added to the model and the model is solved by GAMS. Each product Production quantity on each machine was obtained in a three month horizon. Actual amount production of cold-rolled machines has been compared with the outputs of the model. The results show that by implementing this model, the number of delivered orders and also usage of the machines capacity are increased.

Taking into account the uncertain time duration for each activity in a pert network, we would need to accept the notion that the critical path the project could vary a number of times during its execution. If this variations take place frequently during the project in an unpredictable manner, it could endanger the efficient managment of the project and in addition to lengthening the time frame of execution, it would result in financial burden.The objective of this article is to present an algorithm that potentially prevents the occurance of unpredictable events in the critical path. Using the proposed algorithm, after the identification of risk in each activity and the quantitative and qualitative analyses of the activities,we calculate the respective time durations considering the different events that could carry potential risk for the purpose of risk management. Furthurmore, a parameter called MVC identifies an interval and if the related activity duration falls in that interval, the critical path would be changed.Then through the techniques of Enginerring Economy and Decision Making, conclusions are drawn to whether allow the critical path to undergo variation under anticipated situation or not.

This paper considers a parallel machine scheduling problem with outsourcing allowed. The objective of this problem is the combination of makespan and Outsourcing costs. In order to solve the problem, A mathematical model is proposed. Because of high computational time of mathematical model a Tabu search and PSO methods are proposed to solve the problem.

In each organization, management pursues implementing systems for attaining his objectives until provides suitable environment for presenting services. Call centers are one of the service centers that have been investigated in recent years. The development of call center industry resulted in complicating these systems that has leaded to complicated management and design of these systems. The objective of this paper is the evaluation of call center systems. For this objective, first with the understanding most important problems in call center units, the simulation model of this system is designed and then validity of this model for adapting it with real system is analyzed. At the end, some solutions are presented for improving the performance of system by using design of experiments. The proposed scenarios are simulated and improved model is introduced.

The railroad blocking problem is one of the most important planning problem in freight railways. By solving this problem, once can minimize volume of switching process and total cost of delivering the commodities. This paper presents a method based on branch and price algorithm for solving the railroad blocking problem. This algorithm is a combination form of branch-and-bound and column generation algorithms. Branch and price is a variant of branch and bound, with bounds provided by solving linear programs using column generation at nodes of the branch and bound tree. In branch and price algorithm, re-optimize column generation algorithm in each of branches. Because the bound provided by the LP relaxation is weak, we suggest cuts to strengthen it and show the effect of adding them on the column generation procedure. Implementation of this algorithm has been done by Java programming language. To analyze the quality of the algorithm, some simulated problems with different size generated and are solved by CPLEX software. The results of our algorithm compared with CPLEX’s results. The comparison shows the efficiency and accuracy the purpose algorithm.

Decision Calculus (DC) is a judgment-based approach which considers managers’ opinion in formulation of real world situations. In DC, manager’ mind and the model are aggregated to each other. Little in 1970 introduced the philosophy of decision calculus and it has been used widely in formulation of problems in marketing. The main purpose of this paper is to provide an application for DC in allocation of marketing budgets as a critical activity in marketing. In case of data availability and environmental complexity, researchers suggest the use of mixed models to face such situations to allocate marketing budgets. Mixed models are models which use all the data from the past and managers’ judgment altogether to provide more effective models. In this paper we suggest a simple mixed model. Then we provide a more complex model based on DC and Markov Chains Models (MCM). The suggested models are predicted to be effective in better decision making on allocation of customer relationship budgets.

In recent years, as well as univariate and multivariate statistical process control techniques, profile monitoring has been attracted a great deal of attention of many researchers. A profile is a function between response and independent variables that is monitored during the time. There are some assumptions in different monitoring methods. One of the most important assumptions in most of the procedures is independence of residuals. Violation of this assumption can impress the performance of the most of the methods seriously. Many researchers attempted to eliminate or reduce these effects. In this paper we propose a new method of monitoring of auto correlated linear profile in phase II. The performance of our method has been evaluated by using simulation study and ARL criteria and compared with proposed similar methods.