Journal of Advances in Industrial Engineering
Volume:45 Issue: 1, 2011

In this paper we present a method for optimal design of combined EWMA-EWMA control scheme. Shewhart control chart, introduced by Walter Shewhart, only takes into account present information of the process, so it is insensitive in detecting small shifts. EWMA and CUSUM control charts are good choices for detecting small shifts. A combined control scheme monitors the process for large and small shifts simultaneously. The purpose of this paper is to develop a computer program for optimal design of combined EWMA-EWMA control scheme based on multiple objective optimization problems for detecting large and small shifts simultaneously.

One of the biggest problems of using Data Envelopment Analysis for the evaluation of performance is the weakness of the separability for the decision maker units. This problem is generalizable because of the lower quantity of units in comparison with the ¬input and output quantities of the model [1]. This matter considerably reveals itself in the evaluation process of 23 provincial gas companies considering the higher input and output rates of each provincial gas company. Based on this fact, in this research an integrative model composed of Performance Predictor Neural Networks and Data Envelopment Analysis is designed and used. The results using this model illustrate the strength of the neural network in evaluation and classification of the companies based on their efficiencies.

Demand management is one of the main subjects in marketing which affects the operations performance of an organization considerably. In the classic production planning approach, demand is considered as an exogenous factor in planning, while in the new approach, it is a factor that can be controlled. In this paper, we focus on a new approach in supply chain planning and present an integrated production planning and demand management model is formulated in which demand is controlled by price and market selection. We assume the firm faces a number of potential markets whose demands are price-dependent, and the firm desires to select the best portfolio of markets and the best supply size for each market with regard to his production capacity and markets constraints. An MINLP model is developed for this problem. This model is NP-hard, and so a hybrid heuristic algorithm is developed to solve the problem. A numerical example is also provided to show the applicability of the proposed model.

On time supply of materials in construction projects is very important. Therefore, coordinating a project with key suppliers, while ensuring the timely supply of materials, make the minimum costs of project. One way to create this coordination is integrated inventory planning which has been accomplished in this study. It is necessary to mention that such a model is considered NP_hard in terms of complexity class, because the model is compound of the inventory planning and planning with limited resources, that in literature, the class of programs with limited resources is considered NP_hard. Therefore, in this paper while introducing the inventory planning model in project supply chain, a model based on genetic algorithm is proposed to determine the solution. The proposed algorithm has been run for the suitable size problem and the results show that the introduced algorithm, in actual sizes, within a reasonable time, provides an appropriate response.

Resource Constraints Project Scheduling Problem (RSPSP) seeks proper sequence of implementation of project activities in a way that the precedence relations and different type of resource constraints are met concurrently. RCPSP tends to optimize some measurement function as make-span, cost of implementation, number of tardy tasks and etc. As RCPSP is assumed as an NP-Hard problem so, different meta-heuristic approaches have been proposed to solve different variants of it. In this paper, a modified Ant Colony Optimization (ACO) approach has been developed to deal with RCPSP. The definition of probabilistic selection rule has been modified in proposed approach in favor of better performance. Moreover, the parameters of algorithm have been determined in an adaptive manner and the stagnation behavior has been prevented in high iterations of algorithm. Uncertainty of parameters of RCPSP has also been discussed. The proposed algorithm has been coded using Visual Basic software and tested on benchmark instance in this area. The results are promising and have been compared with optimal or best known solutions.

Nowadays energy has significant role in socio economic development of counties; any decision on energy sector may affect the broad area in societies. System approach in energy modeling as an efficient method is known by many researchers in management and planning in a country particularly in energy sector. Also System Dynamics is one of the efficient methods, for reason unique ability at solving problems toward classic methods that among them consists possibility comprehensive and dynamic investigation, it is possible explain of qualitative variables, definition of different scenario and can be integrated with other method such as econometric method. Hence, in this paper this method has been applied to analyzing effects of energy price changes at household section. At the first, effective factors on energy consumption were identified consider related with household energy demand in literature, that these are such as actual energy price, actual income and number of household, also other attributes effect on these factors and all effective issues are determined by causal loop diagram. Then effects of these factors are have described by related equations, that major equations are explained to clarifying. Also measure of energy saving is subtracted of energy consumption as one of the factors that at next effect on it. In order to validate the proposed model the results of simulation for energy consumption are compared with the actual data for years 2000-2006. Next different scenario analysis for energy price such as considering subsidy, eliminating subsidy, and gradually eliminating subsidy are demonstrated. Then energy consumption and major variables such as energy saving, GDP and labor force are simulated until 2020. In the end, the results of analyzing various scenarios confirm gradually elimination of subsidy (during 10 years) and also moving the subsidies toward building standardization as the concluding remarks of this paper. Also step eliminate subsidy present as an optimum scenario.

Researches prove that effective customer management along with using customer knowledge is one of the topics that can produce competitive advantage in comparison with the other e-commerce topics. Mining the customer knowledge with the aim of helping the managers in better Ad strategy making will be possible using the systems with ability of extracting the behavioral patterns of customers from different databases. The output of these systems can be applied in ad strategy making for e-catalog designs. In this paper, a new framework based on customer knowledge was proposed for e-catalog designs. In this framework, a new algorithm naming DVlw was used for mining the multi-dimensional sequential patterns. The framework was implemented in a computer peripherals and devices Vendor Company. The results show that extracted final knowledge by this framework is more comprehensive comparing with the other frameworks.

An important aspect of customer order supply systems in supplier firms is to manage the processes of "order picking in the warehouses, packing and preparing the shipments for being loaded into the vehicles and transported to the customers". In this regard, two major issues should be considered under resource restrictions: 1) On time customer delivery (meeting deadline) and 2) preparing and shipping the orders with the minimum cost. In this paper, a discrete simulation method is applied to study the processes and analyzing the performance of the customer order supply system of a supplier of spare parts for automobile. Alternative configurations have also been investigated for its operations using time-cost trade-off approach. Nine scenarios were simulated to evaluate the effect of 5, 10, 15, 30, 60, 90, 120, 150 and 180-minute waiting periods (base scenario was 10-minute waiting period), for receiving orders and afterwards planning to both picking them and providing vehicles for delivering the shipments to customers. The data (60000 records) have been collected during a typical period of 118 days of performance of the system under consideration. The study shows that the total shipment cost will be reduced considerably by implementing the optimal scenario. This research is also useful for similar order supply systems in various industries, especially in spare part ones.

Considering the large amount of investment in railway transportation systems, effective use of resources is highly important. Typically, in order to utilize the line capacity, train scheduling plays a substantial role and thus attracts lots of attention all over the world. Various train-scheduling models are introduced, according to different conditions of different countries. However, for Iran, no comprehensive scheduling model has been proposed yet, due to obligatory stops for the passenger trains to say their prayers. In this paper, for first time, a comprehensive mathematical model by considering the constraint of obligatory stops for prayer is presented. The constraint of obligatory stops for prayer is a domestic and religious attribute. In fact, Muslims believe it as essential to pray to God, in the certain times, with special rules or conditions. One of these rules is the restricted period of daily prayer. Another rule of praying is to keep still and calm, meanwhile. Moreover, the person must keep his/her body toward a certain direction, toward the City of Mecca, and should not turn away all during their prayers. All the rules and conditions, above, necessitate stopping of the train in a certain period and in a suitable place. In this paper, the constraint of praying is modeled in three steps: 1. Recognition of the occurrence of praying conditions for each train at each station. 2. Recognition of stopping necessity for praying for each train. 3. Selection of the most appropriate station for stop. At step 1, in order to recognize whether a train will reach a certain station in the permissible time of praying or not, it must confirm that the train arrives at the station after the beginning of this period and before the end of it. Only in the existence of both these conditions, the train can reach to the certain station in the permissible period and therefore can make a stop there. At step 2, it verifies that the train has to stop for praying or not. In order to confirm the necessity or lack of necessity for stop at the intermediate station for praying, three groups of constraints are designed and integrated in the model. The first group of constraints verifies that the passengers have enough time for praying at the origin station. The second group of constraints verifies that the passengers have enough time at the destination station. Finally, the third group of constraints combines the results from the other two groups of constraints and determines in which trains, passengers do not have enough time for praying in either stations of origin and destination and therefore they must make a stop for praying on their way. At step 3, based on consequences from steps 1 and 2, the best station is determined for making a stop on their ways. Finally, a small example and 11 sample problems are solved and the results are presented.