Showing 4 results for Soltani
Mohammad Reza Mehregan, Arman Rezasoltani, Amir Mohammad Khani, Ali Hosseinzade Kashan,
Volume 0, Issue 0 (IN PRESS 2025)
Abstract
In the modern industrial view, it is strategically imperative to predict failure of industrial machinery with a view to reducing the occurrence of unexpected failures and enhancing operational efficiency. This study seeks to introduce a new hybrid machine learning model for predictive maintenance, combining the use of deep learning and advanced ensemble machine learning models. The model presented follows a stacking ensemble structure, where XGBoost, CatBoost, Gradient Boosting, and a deep neural network are base learners. Thereafter, the LightGBM, acting as a meta-model, is used to collect its predictions. Further, in this study, the Optuna hyperparameter optimization framework is employed to optimize the hyperparameters automatically, and the NearMiss algorithm solves the class imbalance problem by enhancing the representation of the minority class and removing the bias in favor of the majority class. As can be seen in the experimental results, the combined model outperforms the single models, achieving an outstanding accuracy of 96.17%. This is followed by a precision of 97.86%, a recall of 94.4%, and an F1 score of 96.1%. It is worth noting that though the XGBoost models' independent results were high (with an F1 score of 89/41) and better than the 16 individual models studied in this paper and regarded as a comparison to the hybrid model, the hybrid model significantly defeated the independent models by nearly 7 percentage points, hence the strong suit of the smart ensemble framework in model combination. The model has been tried using industrial data with 10000 records of a milling machine system, which is representative of most industrial machinery. The model aids in making decisions in preventive maintenance processes in a more informed and timely way by detecting failures accurately before they happen, avoiding unwanted situations of unplanned downtime and operation costs. One can arrive at the conclusion based on these results that the mentioned hybrid model can offer a solid and workable way of predicting failures in the industrial context and can also be integrated into the actual maintenance processes without any issues.
Sajjad Aslani Khiavi, Hamid Khaloozadeh, Fahimeh Soltanian,
Volume 32, Issue 1 (IJIEPR 2021)
Abstract
In this paper, discrete time dynamic model for four-level supply chain system, including factory, wholesaler, retailer, and customer is designed with a recovery center as recycling hybrid channels. Due to the lack of coordination of the chain level and the unhealthy exchange of information in the system, almost all supply chains dynamic involving the stochastic noise. For the first time, in this paper, we proved that stochastic noise lead to the bullwhip effect and we mitigated this phenomenon with control theory. Also, we investigate the effects of the lead time, the various forecasting methods, and aggressive ordering on the bullwhip effect. In order to mitigate the bullwhip effect, we propose Kalman filter method. So, using linear quadratic Gaussian controller, not only effect of bullwhip was adjusted but, also the system become stable. Eventually, the simulation results in Meshkin match factory, indicate the efficiency of the proposed method.
Mostafa Soltani, R. Azizmohammadi, Seyed Mohammad Hassan Hosseini, Mahdi Mohammadi Zanjani,
Volume 32, Issue 2 (IJIEPR 2021)
Abstract
The blood supply chain network is an especial case of the general supply chain network, which starts with the blood donating and ends with patients. Disasters such as earthquakes, floods, storms, and accidents usually event suddenly. Therefore, designing an efficient network for the blood supply chain network at emergencies is one of the most important challenging decisions for related managers. This paper aims to introduce a new blood supply chain network in disasters using the hub location approach. After introducing the last studies in blood supply chain and hub location separately, a new mixed-integer linear programming model based on hub location is presented for intercity transportation. Due to the complexity of this problem, two new methods are developed based on Particle Swarm Optimization and Differential Evolution algorithms to solve practical-sized problems. Real data related to a case study is used to test the developed mathematical model and to investigate the performance of the proposed algorithms. The result approves the accuracy of the new mathematical model and also the good performance of the proposed algorithms in solving the considered problem in real-sized dimensions. The proposed model is applicable considering new variables and operational constraints to more compatibility with reality. However, we considered the maximum possible demand for blood products in the proposed approach and so, lack of investigation of uncertainty conditions in key parameters is one of the most important limitations of this research.
Fatemeh Hajisoltani, Mehdi Seifbarghy, Davar Pishva,
Volume 34, Issue 1 (IJIEPR 2023)
Abstract
The main objective of this research is effective planning as well as greener production and distribution of mineral products in supply chain network. Through a case study in cement industry, it considers the design of the mining supply chain network including several factories with a number of production lines and multiple distribution centers. It leaves part of the transportation operation to contractor companies so as to enable the core company to better focus on its products’ quality and also create job opportunities to local people. It employs a multi-period and multi-product mixed integer linear programming model to both maximize the profit of the factory as well as minimize its carbon dioxide gas emissions which are released during cement production and transportation process. Due to the uncertainty of its cost parameters, fuzzy logic has been used for the modeling and solved via a novel fuzzy multi-choice goal programming approach. Sensitivity analysis has also been done on some key parameters. Comparing results of the model with those from the single-objective models, shows that the model has good efficiency and can be used by managers of mining industries such as cement. Although leaving part of the transportation operations to contractor companies increases the number of vehicles used by the contractor companies, its associated decrease in the number of required factory vehicles, improves both objectives of the model. This should be considered by the managers since on top of profit maximization, it can help them build an eco-friendly image. Mining industries generally generate significant amount of pollutions and companies that pay attention to different dimensions of their social responsibilities can remain stable in the competitive market.
