International Journal of Industrial Engineering & Production Research

---

Decision making about outsourcing or insourcing of manufacturing activities is a type of multiple criteria decision making (MCDM) problem, which requires considering quantitative and qualitative factors as evaluation criteria simultaneously. Therefore, a suitable MCDM method can be useful in this area as it can consider the interactions among quantitative and qualitative criteria. The analytic network process (ANP) is a relatively new MCDM method which can deal with different kinds of interactions systematically. Moreover, the Decision Making Trial and Evaluation Laboratory (DEMATEL) method is able to convert the relations between cause and effect of criteria into a visual structural model as well as handling the inner dependences within a set of criteria. However both ANP method and DEMATEL techniques in their original forms are incapable of capturing the uncertainty during value judgment elicitation. To overcome this problem, here, a new and effective model is proposed based on combining fuzzy ANP and fuzzy DEMATEL for decision making about outsourcing or insourcing of manufacturing activities in uncertain conditions. Data from a case study is used to illustrate the usefulness and applicably of the proposed method.

---

The analytic center cutting plane method (ACCPM) is one of successful methods to solve nondifferentiable optimization problems. In this paper ACCPM is used for the first time in the vehicle routing problem with time windows (VRPTW) to accelerate lagrangian relaxation procedure for the problem. At first the basic cutting plane algorithm and its relationship with column generation method is clarified then the new method based on ACCPM is proposed as a stabilization technique of column generation (lagrangian relaxation). Both approaches are tested on a benchmark instance to demonstrate the advantages of proposed method in terms of computational time and lower bounds quality.

---

This paper presents a new formulation for warehouse inventory management in a stochastic situation. The primary source of this formulation is derived from FP model, which has been proposed by Fletcher and Ponnambalam for reservoir management. The new proposed mathematical model is based on the first and the second moments of storage as a stochastic variable. Using this model, the expected value of storage, the variance of storage, and the optimal ordering policies are determined. Moreover, the probability of within containment, surplus, and shortage are computable without adding any new variables. To validate the optimization model, a Monte Carlo simulation is used. Furthermore, to evaluate the performance of the optimal FP policy, It is compared to (s*,S*) policy, as a very popular policy used in the literature, in terms of the expected total annual cost and the service level. It is also demonstrated that the FP policy has a superior performances than (s*,S*) policy.

---

Increasing competition in the air transport market has intensified active airlines’ efforts to keep their market share by attaching due importance to cost management aimed at reduced final prices. Crew costs are second only to fuel costs on the cost list of airline companies. So, this paper attempts to investigate the cockpit crew pairing problem. The set partitioning problem has been used for modelling the problem at hand and, because it is classified in large scale problems, the column generation approach has been used to solve LP relaxation of the set partitioning model. Our focus will be on solving the column generation sub-problem. For this purpose, two algorithms, named SPRCF and SPRCD, have been developed based on the shortest path with resource constraint algorithms. Their efficiency in solving some problem instances has been tested and the results have been compared with those of an algorithm for crew pairing problem reported in the literature. Results indicate the high efficiency of the proposed algorithms in solving problem instances with up to 632 flight legs in a reasonable time.

---

In the fundamental production inventory model, in order to solve the economic production quantity (EPQ) we always fix both the demand quantity and the production quantity per day. But, in the real situation, both of them probably will have little disturbances every day. Therefore, we should fuzzify both of them to solve the economic production quantity (q*) per cycle. Using α-cut for defuzzification the total variable cost per unit time is derived. Therefore the problem is reduced to crisp annual costs. The multi-objective model is solved by Global Criteria Method with the help of GRG (Generalized Reduced Gradient) Technique. In this model shortages are permitted and fully backordered. The purpose of this paper is to investigate a computing schema for the EPQ in the fuzzy sense. We find that, after defuzzification, the total cost in fuzzy model is less than in the crisp model. So it permits better use of the EPQ model in the fuzzy sense arising with little disturbances in the production, and demand.

---

The set covering problem (SCP) is a well-known combinatorial optimization problem. This paper investigates development of a local branching approach for the SCP. This solution strategy is exact in nature, though it is designed to improve the heuristic behavior of the mixed integer programming solver. The algorithm parameters are tuned by design of experiments approach. The proposed method is tested on the several standard instances. The results show that the algorithm outperforms the best heuristic approaches found in the literature.

---

Considering the major change occurred in business cells from plant to “chain” and the critical need to choose the best partners to form the supply chain for competing in today’s business setting, one of the vital decisions made at the early steps of constructing a business is supplier selection. Given the fact that the early decisions are inherently strategic and therefore hard and costly to change, it’s been a point of consideration for industries to select the right supplier. It’s clear that different criteria must be investigated and interfered in deciding on the best partner(s) among the alternatives. Thereupon the problem might be regarded as a multiple criteria decision making (MCDM) problem. There are a variety of techniques to solve a MCDM problem. In this paper we propose a novel technique by combination of decision making trial and evaluation laboratory and graph theory and matrix approach techniques. Eventually, the results are compared to SAW technique and discussed to come to a conclusion.

---

Nowadays, in turbulent and violate global markets an Agile Supply Chain (ASC) network is frequently considered as a dominant competitive advantage for survival. To achieve the competitive advantage, companies must align with suppliers and customers to streamline operations, as well as agility beyond individual companies. There are many definitions and models about agile supply chain and most of them have emphasized on capabilities and enablers, and their sub attributes, as two critical factors, but regardless of time and its effect on the main attributes. This paper tries to present the role of time on predicting the agility of supply chain, by studying effect of intervention time on enablers and eventually predict the progress trend of agility in supply chain. To gain this end we use ANFIS output to assess agility and compare the effects of agile enablers in period of time on capabilities in Iran Khodro manufacture. This recognition helps managers to consider time as leverage factor and focus on this factor to enhance existent agility level and achievement the desired one.

---

Data envelopment analysis operates as a tool for appraising the relative efficiency of a set of homogenous decision making units. This methodology is applied widely in different contexts. Regarding to its logic, DEA allows each DMU to take its optimal weight in comparison with other DMUs while a similar condition is considered for other units. This feature is a bilabial characteristic which optimizes the performance of units in one hand. This flexibility on the other hand threats the comparability of different units because different weighting schemes are used for different DMUs. This paper proposes a unified model for determination of a common set of weights to calculate DMUs efficiency. This model is developed based on a multi objective fractional linear programming model that considers the original DEA's results as ideal solution and seeks a set of common weights that rank the DMUs and increase the model's discrimination power. Comparison of the proposed method with some of the previously presented models has shown its advantages as a DMUs ranking model.

---

In this paper we develop a new approach for land leveling in order to improve the topology of a large area for irrigation or civil projects. The objective in proposed model is to minimize the total volume of cutting so that technical requirements of land leveling such as suitable slope and standard ratio of cutting to filling and maximum penstock point’s height are considered. We develop a warped surface pattern and apply a linear programming model to determine the land optimal topology. Our approach is more practical to apply, in comparison with the existing “fit to plane” methods which apply bivariate regression statistical techniques because in these methods finding optimal solution, considering technical requirements, needs trial and error. Our proposed method does not need any trial and error, furthermore its results is global optimum. Also the warped surface pattern is adoptable to plane or curved patterns, and it is applicable for any land with any magnitude.

---

Disasters can cause many casualties and considerable destruction mainly because of ineffective preventive measures, incomplete preparedness, and weak relief logistics systems. After catastrophic events happen, quick and effective response is of great importance, so as to having an efficient logistic plan for distributing needed relief commodities efficiently and fairly among affected people. In this paper, we propose a fuzzy multi-objective, multi-modal, multi-commodity logistic model in emergency response to disaster occurrence, to assign limited resources equitably to the infected regions in a way to minimize transfer costs of commodities as well as distribution centers activation costs, and maximizing satisfied demand. In the proposed model, we have determined the optimal place of distribution centers among candidate points to receive people donations as well as sending and receiving different kinds of relief commodities. The amount of voluntary donations is not known precisely and is estimated with uncertainty, so we have used fuzzy parameters for them. The number of victims immediately after disaster is vague and is estimated indecisively though we have considered it as a fuzzy demand. A case study has been displayed to test the properties of the optimization problem that shows efficiency of this formulation in experiment.

---

Pricing and advertising are two important marketing strategies in the supply chain management which lead to customer demand’s increase and therefore higher profit for members of supply chains. This paper considers advertising, and pricing decisions simultaneously for a three-level supply chain with one supplier, one manufacturer and one retailer. The amount of market demand is influenced by pricing and advertising. In this paper, three well-known approaches in the game theory including the Nash, Stackelberg and Cooperative games are exploited to study the effects of pricing and advertising decisions on the supply chain. Using these approaches, we identify optimal decisions in each case for the supplier, the manufacturer and the retailer. Also, we compare the outcomes decisions among the mentioned games. The results show that, the Cooperative and the Nash games have the highest and lowest advertising expenditure, respectively. The price level in the Nash game is more than the Stackelberg game for all three levels, and the retailer price in the Stackelberg and Cooperative games are equal. The system has the highest profit in the Cooperative game. Finally, the Nash bargaining model will be presented and explored to investigate the possibilities for profit sharing.

---

The locomotive assignment and the freight train scheduling are important problems in railway transportation. Freight cars are coupled to form a freight rake. The freight rake becomes a train when a locomotive is coupled to it. The locomotive assignment problem assigns locomotives to a set of freight rakes in a way that, with minimum locomotive deadheading time, rake coupling delay and locomotive coupling delay all freight rakes are hauled to their destinations. Scheduling freight trains consists of sequencing and ordering freight trains during the non-usage time between passenger trains but with no interference and with minimum delay times. Solving these two problems simultaneously is of high importance and can be highly effective in decreasing costs for rail transportation. In this paper, we aim to minimize the operational costs for the locomotive assignment and the freight train scheduling by solving these two problems concurrently. To meet this objective, an efficient and effective algorithm based on the ant colony system is proposed. To evaluate the performance of the proposed solution method, twenty-five test problems, which are based on the conditions of Iran Railways, are solved and the computational results are reported.

---

In the current study, a dual-channel supply chain is considered containing one manufacturer and two retailers. It is assumed that the manufacturer and retailers have the same decision powers. A game-theoretic approach is developed to analyze pricing decisions under the centralized and decentralized scenarios. First, the Nash model is established to obtain the equilibrium decisions in the decentralized case. Then, the centralized model is developed to maximize the total profit of the whole system. Finally, the equilibrium decisions are discussed and some managerial insights are revealed. 

---

The present paper aims to propose a fuzzy multi-objective model to allocate order to supplier in uncertainty conditions and for multi-period, multi-source, and multi-product problems at two levels with wastages considerations.  The cost including the purchase, transportation, and ordering costs, timely delivering or deference shipment quality or wastages which are amongst major quality aspects, partial coverage of suppliers in respect of distance and finally, suppliers weights which make the products orders more realistic are considered as the measures to evaluate the suppliers in the proposed model. Supplier's weights in the fifth objective function are obtained using fuzzy TOPSIS technique. Coverage and wastes parameters in this model are considered as random triangular fuzzy number. Multi-objective imperial competitive optimization (MOICA) algorithm has been used to solve the model,. To demonstrate applicability of MOICA, we applied non-dominated sorting genetic algorithm (NSGA-II). Taguchi technique is executed to tune the parameters of both algorithms and results are analyzed using quantitative criteria and performing parametric. 

---

This paper addresses the Tardy/Lost penalty minimization on a single machine. According to this penalty criterion, if the tardiness of a job exceeds a predefined value, the job will be lost and penalized by a fixed value. Besides its application in real world problems, Tardy/Lost measure is a general form for popular objective functions like weighted tardiness, late work and tardiness with rejection and hence, the results of this study are applicable for them. Initially, we present two approximation algorithms. Then, two special cases of the main problem are considered. In the first case, all jobs have the same tardiness weights where an FPTAS is developed using the technique of “structuring the execution of an algorithm". The second special case occurs when none of the jobs can be early. For this case, a 2-approximation algorithm is developed as well as a dynamic programming algorithm which is converted to an FPTAS.

---

Considering the high costs of the implementation and maintenance of gas distribution networks in urban areas, optimal design of such networks is vital. Today, urban gas networks are implemented within a tree structure. These networks receive gas from City Gate Stations (CGS) and deliver it to the consumers. This study presents a comprehensive model based on Mixed Integer Nonlinear Programming (MINLP) for the design of urban gas networks taking into account topological limitations, gas pressure and velocity limitations and environmental limitations. An Ant Colony Optimization (ACO) algorithm is presented for solving the problem and the results obtained by an implementation of ACO algorithm are compared with the ones obtained through an iterative method to demonstrate the efficiency of ACO algorithm. A case study of a real situation (gas distribution in Kelardasht, Iran) affirms the efficacy of the proposed approach.
 

---

Abstract
Production and distribution network (PDN) planning in multi-stage status is commonly complex. These conditions cause significant amount of uncertainty relating to demand and lead time. In this study, we introduce a PDN to deliver the products to customers in the least time and optimize the total cost of the network, simultaneously. The proposed network is four stage PDN including suppliers, producers, potential entrepots, retailers and customers with multi time period horizon with allowable shortage. A mixed integer programming model with minimizing total cost of the system and minimizing total delivery lead time is designed. We present a novel heuristic method called selective firefly algorithm (SFA) in order to solve several sized especially real world instances. In SFA, each firefly recognizes all better fireflies with more brightness and analyses its brightness change before moving, tacitly. Then, the firefly that makes best change is selected and initial firefly moves toward the selected firefly. Finally, the performance of the proposed algorithm is examined with solving several sized instances. The results indicate the adequate performance of the proposed algorithm.

---

This paper deals with multi- objective nonlinear programming problem having rough intervals in the constraints. The problem is approached by taking maximum value range and minimum value range inequalities as constraints conditions, reduces it into two classical multi-objective nonlinear programming problems, called lower and upper approximation problems.  All of the lower and upper approximation problems may be solved using the weighting method, where an optimal rough interval solution is obtained. The stability set of the first kind corresponding to the optimal rough interval solution is determined. An illustrative numerical example is given to clarify the obtained results.

---

     Fully fuzzy linear programming is applied to water resources management due to its close connection with human life, which is considered to be of great importance. This paper investigates the decision-making concerning water resources management under uncertainty based on two-stage stochastic fuzzy linear programming. A solution method for solving the problem with fuzziness in relations is suggested to prove its applicability. The purpose of the method is to generate a set of solutions for water resources planning that helps the decision-maker make a tradeoff between economic efficiency and risk violation of the constraints. Finally, a numerical example is given and is approached by the proposed method.