Scientia Iranica
Volume:28 Issue: 1, Jan-Feb 2021

In order to fully excavate the information contained in the multi-index panel data, one take decision objects as the research object, and the development state matrix and the development speed matrix of the decision objects are defined by considering the cross-section information and time information of the decision objects, and then the distances among the objects over the indexes are given. Based on grey incidence analysis, the absolute difference and relative difference between the measure value matrices are used to characterize and measure the close degree of the development state matrix and the development level matrix of the decision objects, so that the grey object matrix absolute incidence analysis model is established, and then according to the grey incidence degree between the objects, the objects can be clustered based on hierarchical clustering algorithm. Finally, a clustering problem of regional patent research and development (R&D) efficiency is used to verify the validity and rationality of the proposed model.

The aim of the paper is to develop a grey model for short term forecasting of natural gas consumption and production in China and USA respectively. To enhance its prediction accuracy, the outliers are found by the error of the latent information function, and then corrected according to the test sample and the future trend. The sequence with corrected outliers is used to construct a grey model. The proposed model is employed to predict the natural gas consumption and production in China and USA. The results have demonstrated that the proposed model can raise the forecast accuracy of the grey model, and it also indicates that China will inevitably face a massive expansion in natural gas imports.

The use of artificial intelligence-based models have shown that the market is predictable despite its uncertainty and unstable nature. The most important challenge of the proposed models in the stock market is the accuracy of the results and increasing the forecasting efficiency. Another challenge, which is a prerequisite for making decision and using the results of the forecast for profitability of transactions, is to forecast the trend of stock price movements in forecasting price. To overcome the mentioned challenges, this paper employs ensemble learning (EL) model using intelligence-based learners and metaheuristic optimization methods to maximize the improvement of forecasting performance. In addition, in order to consider the direction of price change in stock price forecasting, a two-stage structure is used. In the first stage, the next movement of the stock price (increase or decrease) is forecasted and its outcome is then employed to forecast the price in the second stage. In both stages, genetic algorithm (GA) and particle swarm optimization (PSO) technique are used to optimize the aggregation results of the base learners. The evaluation results of stock market dataset show that the proposed model has higher accuracy compared to other models used in the literature.

Facility location problem is a branch of operational research and computational geometry. It involves the best allocation of facilities to minimize transportation costs, while considering factors such as avoiding placing dangerous materials near the premises and the facilities of competitors. According to B2C e-commerce unique customer characteristics and fierce market competition, two facility location models in e-commerce under uncertainty are proposed, i.e., expected value model and pessimistic value model. It is proved these models can be converted into equivalent models based on inverse uncertainty distribution method. A hybrid algorithm is proposed to solve these models. Some numerical experiments are used to demonstrate the effectiveness of the proposed models and approach.

In this paper, we considered the competition between two multi-echelon supply chains with identical structures on price under two market power structures. For this purpose, we developed two different scenarios. In the first scenario, both supply chains decided simultaneously (the Nash game). In the second scenario, due to the imbalance of power between the two supply chains, we adopted the Stackelberg game in the model. Price equilibrium is obviously obtained through the Game Theory. The paper investigated the effects of different relations between the market sizes of supply chains and the supply chain structures on price and profit along with the analysis of power in the market. Based on these assumptions, it was found that the supply chains did not always involve the second-mover advantage in the price Stackelberg game. Furthermore, having the centralized structure, both of the supply chains benefited from presence of a leader in the market for different combinations of market size. Finally, we presented significant managerial insights for the market with two competitive supply chains when the structures were similar. Moreover, the relationship between price and profit was analyzed given the size of the market in different scenarios rather than through provision of numerical examples.

In the field of supply chain, selecting a suitable green supplier could significantly help us to decrease the cost and the risk involved in the operations as well as increase in the quality and green. In this paper, we develop an integrated two-stage approach based on fuzzy analytic hierarchy process (FAHP) and multi-objective mixed-integer linear programming to select suppliers and order allocation in green supply chain. In the first stage, suppliers are evaluated using FAHP method, and in the second stage, a multi-product multi-period supply chain considering green location-routing problem, discounting, and time window under uncertainty is developed. Then, a fuzzy solution approach is applied to solve proposed model using the data of a pharmaceutical chain in Iran. Results will verify the efficiency of the proposed model.

Utilization of superpopulation models for estimation of population parameters is an advantageous practice, when it is easy to recognize the relationship between the study variable with one or more auxiliary variables. This article is concerned with estimation of finite population total under a new ranked set sampling approach, ranked set sampling without replacement (RSSWOR), using so called gamma population model (GPM). Behavior of the proposed estimator, in term of relative efficiency, is studied for various choices of a constant γ via Monte Carle experiment. The provided simulation study shows the superiority of the proposed estimator over existing estimator under same model. The sampling procedure, especially, aids in collecting data from a continuous production process.

The quality of public transportation service has major effects on people’s quality of life. During frequency and timetable setting, synchronization is a very important and complicated issue which can directly influence the utility and attractiveness of the system. In this paper, a mixed-integer nonlinear programming model is proposed that aims at setting timetables on a bus transit network with the maximum synchronization and the minimum number of fleet size. The proposed model is shown to be applicable for both small and large-scale transit networks by employing it for setting timetables on two samples of both sizes. As an illustrative example, a simple version of the model is coded and run in GAMS Software and a completely reasonable timetable is obtained. As the second example, the proposed model is used to set timetables on Tehran BRT networks through the genetic algorithm; then the NSGA-II is used to obtain the Pareto optimal solutions of the problem for five different scenarios. The Pareto optimal solutions are used to draw the Pareto optimal fronts which act as an essential decision making tool. The overall results show that the proposed model is efficient enough to be employed setting timetables on transit networks with different sizes.

In this paper, a three-stage network with optimal desirable and undesirable inputs and outputs has been taken into consideration by us. This network comprises of a leader and two followers. Four diverse models of Data Envelopment Analysis (DEA) to measure the efficiency or the performance, of this three-stage network have been taken under contemplation; these are namely, a Black Box Model and three Stackelberg Game (Theory) Models. A multiplicative DEA, with a double-frontier approach, to measure the efficiency of the entire system and the performances of the decision making units (DMUs), from both the optimistic and pessimistic views have been utilized. In this paper attempts have been made to present the goals of the managers in the models. Hence, aspects of goal programming have been manipulated so as to define cooperation between the leader and followers, such that, we are able to include the objectives of the managers in the models. In actual fact, a non-cooperative collaboration is deliberated upon. In addition to which, in the second and third scenarios, the leader-follower, nonlinear models are present. Thereby, a heuristic approach is suggested to convert the nonlinear models into linear ones.

Eigen fuzzy set of a fuzzy relation often occurs to be invariant under different computational aspects. The present communication introduces the novel concept of eigen spherical fuzzy set of spherical fuzzy relation along with various composition operators for the first time. We have proposed two algorithms to determine the greatest eigen spherical fuzzy sets and least eigen spherical fuzzy sets using the $max-min$ and $min-max$ composition operators respectively and illustrated the steps with the help of flow charts. Further, two numerical examples related to different fields of decision-making problems have been taken into account for illustrating the proposed methodology. The scope of future work in the field of image information retrieval, genetic algorithm for image reconstruction and notion of eigen spherical fuzzy soft sets/matrices has been duly outlined. The comparative remarks and advantages of the proposed eigen spherical fuzzy sets have also been included for a better readability.

Recent years have envisaged a great deal of interest in optimizing of logistics and transforming systems. One of important challenges in this regard is the cross dock scheduling with several real-life limitations such as the deadline for both perishable and imperishable products. This study is a new cross-dock scheduling problem by not only considering a time window but also for all shipping trucks, the deadline is assumed by the presence of perishable products for the first time in this research area. Based on these suppositions, a new mathematical model is developed. The last but not the least is to propose a new hybrid metaheuristic by combining a recent nature-inspired metaheuristic called Keshtel Algorithm (KA) and a well-known algorithm named Simulated Annealing (SA). The proposed hybrid algorithm not only is compared with its individual ones but also some other well-known metaheuristic algorithms are used. Finally, the performance of the proposed algorithm is validated by several experiments with different complexities and statistical analyses.

Control chart (CC) is used to monitor the special causes that arise during the process monitoring. These special causes produce continual shifts in the process parameters that last until it is identified and removed. There is a need for such techniques, which present the true representation of the entire process. Rational subgrouping is an essential concept in Statistical Process Control (SPC) which is seldom overlooked by the practitioner. Hence, most of the manufacturing, engineering, and production processes give output products in the form of batches over smaller intervals of time. The aim of this study is to provide a median based design for Tukey and Tukey-EWMA control charts under subgrouping. It will use the idea of boxplot to monitor the process behavior. This study also provides a brief discussion regarding selecting and forming subgroups from the process data. The performance of the median based Tukey and Tukey-EWMA charts are judged using Average, Median and Standard-Deviation run-length as performance measures. We have considered subgroup sizes of m=1,5 &10 at pre-specified ARL0 equal to 370. To real-life applications of the median based tukey designs are also presented to show their implementation in food manufacturing and hard-bake processes.

Various control charts have been proposed to monitor generalized linear profiles in Phase II. However, robustness of the proposed methods in detecting different types and especially different directions of changes is not well-studied in the literature. In real-world applications different kinds of changes such as drift and multiple change are likely to happen which can be isotonic (increasing) or antitonic (decreasing). This paper studies the robustness of Rao Score Test (RST) method, T2, and multivariate exponential weighted moving average (MEWMA) in different types, drift and multiple, and directions of changes. Rao Score Test method also benefits from a change-point detection approach whose performance is studied as well. According to the results, generally RST method shows a better performance in detecting different types of changes. Moreover, the performance of the RST method is robust to direction of the change, while T2 and MEWMA are not ARL-unbiased and show different performances under isotonic and antitonic changes. Therefore, to address this issue, we proposed a bias-reduced estimator to be used in T2. Our results demonstrate that the proposed control chart outperforms T2 and is less biased than T2. Finally, a real-world problem is presented in which aforementioned methods are applied to real data.