فهرست مطالب mehdi bijari

In this research, lot-sizing and scheduling problem on parallel machines has been studied. Holding inventory and backlog cost has been considered as an earliness-tardiness penalties. A mixed integer programming formulation has been proposed based on TSP. Number of product batch is calculated as a parameter before solving the model. The computational result demonstrated that the MIP uses large CPU time to get result due to the problem complexity. So in the next step, problem has been modeled by constraint programming method that reduces solving time significantly. So that for an instance with 2 hours CPU solving time in MIP, the CP method reduces solving time to 2 minutes. To complete the solving process, a heuristic algorithm is proposed to assign orders to products. A case-study in steel-mill industry shows the efficiency of designed system rather than the existing system. Experimental results show that the proposed system have planned the orders less than 10 minutes solving time for different instances; while this is 1 to 2 hours for the existing system.

In competitive markets, attracting potential customers and keeping current customers is a survival condition for each company. So, paying attention to the requests of customers is important and vital. In this paper, the problem of order acceptance and scheduling has been studied, in which two types of customers or agents compete in a single machine environment. The objective is maximizing sum of the total profit of first agent's accepted orders and the total revenue of second agent. Therefore, only the first agent has penalty and its penalty function is lateness and the second agent's orders have a common due date and this agent does not accept any tardy order. To solve the problem, a mathematical programming, a heuristic algorithm and a pseudo-polynomial dynamic programming algorithm are proposed. Computational results confirm the ability of solving all problem instances up to 70 orders size optimally and also 93.12% of problem instances up to 150 orders size by dynamic programming.

Considering its application, stochastic lot sizing is a significant subject in production planning. Also the concept of service level is more applicable than shortage cost from manager's viewpoint. In this paper, the stochastic multi period multi item capacitated lot sizing problem has been investigated considering service level constraint. First, the single item model has been developed considering service level and with no capacity constraint and then, it has been solved using dynamic programming algorithm and the optimal solution has been derived. Then the model has been generalized to multi item problem with capacity constraint. The stochastic multi period multi item capacitated lot sizing problem is NP-Hard, hence the model could not be solved by exact optimization approaches. Therefore, simulated annealing method has been applied for solving the problem. Finally, in order to evaluate the efficiency of the model, low level criterion has been used.

In recent years, various time series models have been proposed for financial markets forecasting. In each case, the accuracy of time series forecasting models are fundamental to make decision and hence the research for improving the effectiveness of forecasting models have been curried on. Many researchers have compared different time series models together in order to determine more efficient once in financial markets. In this paper, the performance of four interval time series models including autoregressive integrated moving average (ARIMA), fuzzy autoregressive integrated moving average (FARIMA), hybrid ANNs and fuzzy (FANN) and Improved FARIMA models are compared together. Empirical results of exchange rate forecasting indicate that the FANN model is more satisfactory than other those models. Therefore, it can be a suitable alternative model for interval forecasting of financial time series.

This paper breaks new ground by modelling lot sizing and scheduling in a flexible flow line (FFL)simultaneously instead of separately. This problem, called the ‘General Lot sizing and Scheduling Problem in a Flexible Flow Line’ (GLSP-FFL), optimizes the lot sizing and scheduling of multiple products at multiple stages, each stage having multiple machines in parallel. The objective is to satisfy varying demand over a finite planning horizon with minimal inventory, backorder and production setup costs. The problem is complex as any product can be processed on any machine but with different process rates and sequence-dependent setup times & costs. The efficiency of two alternative models is ssessed and evaluated using numerical tests.

Computational intelligence approaches have gradually established themselves as a popular tool for forecasting the complicated financial markets. Forecasting accuracy is one of the most important features of forecasting models; hence, never has research directed at improving upon the effectiveness of time series models stopped. Nowadays, despite the numerous time series forecasting models proposed in several past decades, it is widely recognized that exchange rates are extremely difficult to forecast. Artificial Neural Networks (ANNs) are one of the most accurate and widely used forecasting models that have been successfully applied for exchange rate forecasting. In this paper, a hybrid model is proposed based on the basic concepts of artificial neural networks in order to yield more accurate results than the traditional ANNs in short span of time situations. Three exchange rate data sets—the British pound, the United States dollar, and the Euro against the Iran rial-are used in order to demonstrate the appropriateness and effectiveness of the proposed model. Empirical results of exchange rate forecasting indicate that hybrid model is generally better than artificial neural networks and other models presented for exchange rate forecasting, in cases where inadequate historical data are available. Therefore, our proposed model can be a suitable alternative model for financial markets to achieve greater forecasting accuracy, especially in incomplete data situations.

Both theoretical and empirical findings have suggested that combining different models can be an effective way to improve the predictive performance of each individual model. It is especially occurred when the models in the ensemble are quite different. Hybrid techniques that decompose a time series into its linear and nonlinear components are one of the most important kinds of the hybrid models for time series forecasting. Several researches in the literature have been shown that these models can outperform single models. In this paper, the predictive capabilities of three different models in which the autoregressive integrated moving average (ARIMA) as linear model is combined to the multilayer perceptron (MLP) as nonlinear model, are compared together for time series forecasting. These models are including the Zhang’s hybrid ANNs/ARIMA, artificial neural network (p,d,q), and generalized hybrid ANNs/ARIMA models. The empirical results with three well-known real data sets indicate that all of these methodologies can be effective ways to improve forecasting accuracy achieved by either of components used separately. However, the generalized hybrid ANNs/ARIMA model is more accurate and performs significantly better than other aforementioned models.

The evolution of financial data shows a high degree of volatility of the series, coupled with increasing difficulties of forecasting financial variables. Some alternative forecasting methods, based on the literature review, have been developed, which can be particularly useful in the analysis of financial time series. Despite of the numerous time series forecasting models, the accuracy of time series forecasting is fundamental to many decision processes. Selecting an efficient technique in unique situations is very difficult task for forecasters. Many researchers have integrated linear and nonlinear methods in order to yield more accurate results. In practice, it is difficult to determine the time series under study are generated from a linear or nonlinear underlying process while many aspects of economic behavior may not be pure linear or nonlinear. Although both ARIMA and Artificial Neural Networks (ANNs) models have the flexibility in modeling a variety of problems, none of which is universally the best model used indiscriminately in every forecasting situation. In this paper, based on the foundations of ARIMA and ANNs models, a hybrid method is proposed to forecast exchange rate. Empirical results indicate that integrating linear and nonlinear ARIMA and Artificial Neural Networks (ANNs) models can be an effective way to improve forecasting accuracy achieved by either of the above linear and nonlinear models used separately.