Search

Due to the complex nature of processing units and the variability in raw oil properties, traditional first-principle methods are not always effective in predicting the output of these units without adapting them. Data-driven models have become popular alternatives because they are more adaptable to the variability in processing units, especially when large amounts of data are available. Feature selection is a crucial aspect of data-driven modeling, especially for processing units. It involves identifying and selecting the most important variables influencing the unit's output. The selected variables are then used as inputs to the model, which helps to reduce noise and improve model accuracy. Accordingly, this article presents a hybrid method that combines feature selection with data-driven modeling to create an accurate model for predicting the output of a Residue Fluid Catalytic Cracking (RFCC) process unit. The Metaheuristic Ant Colony Optimizer (MACO) algorithm was used to extract the best subset of features from the collected data, which were then used with the Artificial Immune Genetic Algorithm with Local Vaccination (AIGA-LV) method to select the optimal features. Finally, the Random Forest (RF) method was used to predict the products of the RFCC process unit. The data were collected from a refinery plant in central Iran for nine months. The proposed model was efficient in feature selection and flexible in handling different scenarios and products. It improved the accuracy to 99.17% at its best and improved the primary model by 3–20%. Therefore, this method could be useful for predicting the output of various industrial processes, especially the RFCC and FCC process units. [ABSTRACT FROM AUTHOR]

One of the basic steps in water resources management and planning according to population increase and lack of water resources in Iran is to optimize the use of dam reservoirs. In this research, the effect of meteorological droughts on the optimization of the Aydoghmoush dam reservoir in the northwest of Iran was evaluated by applying metaheuristic algorithms under the impact of future climate change. Three models and two scenarios of SSP2-4.5 and SSP2-8.5 of the sixth IPCC report, and the LARS-WG downscaling model were used for Aydoghmoush dam weather station for the base period (1978-2014) and future periods of 2022-2040 and 2070-2100. The inflow and outflow of the dam, as well as the optimal utilization of the dam reservoir, were evaluated using standalone, and hybrid mode of genetic, slime mold, and ant colony algorithms. Results of the best release scenario (SSP2-8.5) showed that the annual rainfall in the future periods will decrease by 8.9 mm, and 14.5 mm, respectively, compared to the base period. The objective function of optimizing the use of the dam reservoir was defined as minimizing the sum of squared relative deficiencies in each month and maximizing the reliability in the statistical period of 2011-2021. The results showed that in terms of time reliability, vulnerability, and stability, the hybrid slime mold-genetic algorithm was better than other algorithms with values of 0.73, 0.32, and 28.78. Prediction of the dam's inflow and outflow using the hybrid slime mold-genetic algorithm indicated high accuracy compared to other models by 13% and 19% errors, respectively. [ABSTRACT FROM AUTHOR]

Purpose: This paper aims to design and regulate the Adaptive Neural Fuzzy Inference Systems (ANFIS) of supplier development, determine the importance of different categories of supplier development (SD), and suggest appropriate activities based on the results for the development of suppliers in the automotive industry. Design/methodology/approach: To design ANFIS, Grid Partitioning, Subtractive Clustering, and FCM have been used. Then, they have been regulated using Back Propagation (BP), Hybrid, Ant Colony Optimization (ACO), Differential Evolution (DE), Genetic Algorithm (GA), and Particle Swarm Optimization (PSO) methods. A five-step procedure has been used for sensitivity analysis of the supplier development categories. By comparing the results of the regulated systems, the most appropriate ones were selected. Also, based on the supplier development score in different categories separately, the supplier development score was predicted for 53 strategic suppliers in the automotive industry. By sensitivity analysis, SD-related categories were prioritized to guide automotive industry manufacturers to use SD-related activities. Findings: Findings indicated that the FCM compared to the other two methods; meta-heuristic regulation methods compared to BP and Hybrid, and ACO, DE, and GA compared to PSO led to better results. The ACO in all systems, the DE in four systems, and the GA in two categories were identified as the dominant methods, while the PSO was not dominant in any of the categories. This finding implies the priority of meta-heuristic algorithms as ACO> DE> GA> PSO, based on the data of this study. The results of the correlation between the scores of the ANFIS and the average scores of the experts show that the designed ANFIS has high accuracy. Practical implications: The results of this study will direct manufacturers' investments and direct involvement in SD. The findings encourage manufacturers and suppliers of the Iranian automotive industry first to activities related to the development of environmental capabilities and then activities related to the development of three other categories. The managers of Iran's automotive industry are suggested to apply the activities related to the development of suppliers' environmental capabilities. These activities include evaluating the supplier's environmental performance and feedback, sharing environmental information, ethics and social responsibility, obtaining environmental and social certifications, developing programs to improve the quality of life of target communities, green procurement, and environmental awareness, logistics activities Inversion, and joint efforts to improve performance are sustainable. Originality/value: This study was one of the first in-house studies to compare the results of metaheuristic algorithms compatible with ANFIS in the field of SD. In addition, in terms of implementation, it offered suitable SD activities to car manufacturers. [ABSTRACT FROM AUTHOR]

Due to climate change and urban growth, the demand for new freshwater sources, especially groundwater, is increasing in water-deficient countries like Iran. Therefore, this study aimed at groundwater potential mapping (GPM) of the Nahavand Plain, Iran, using an optimized adaptive neuro fuzzy inference system (ANFIS) in a geographic information system, with three metaheuristic optimization algorithms: differential evolution (DE), particle swarm optimization (PSO) and ant colony optimization (ACO). A spatial database was constructed using 273 spring locations and 14 groundwater conditioning factors. The optimization algorithms were evaluated using the receiver operating characteristic (ROC) technique. The ANFIS-DE, ANFIS-PSO and ANFIS-ACO models resulted in accuracy of 0.816, 0.809 and 0.758, respectively; the high and very high potential for groundwater springs covered 26% of the Nahavand Plain. The Root Mean Square Error (RMSE) for the training and validation datasets was lowest for the ANFIS-DE model compared to the other two models; and the ANFIS-PSO model had a higher convergence speed. These results may play an important role in sustainable groundwater management in the Nahavand Plain. [ABSTRACT FROM AUTHOR]

Groundwater is a crucial source of water supply in drought conditions, and an auxiliary water source in wet seasons. Due to its increasing importance in view of climate change, predicting groundwater level (GWL) needs to be improved to enhance management. We used adaptive neuro-fuzzy inference systems (ANFIS) to predict the GWL of the Urmia aquifer in northwestern Iran under various input scenarios using precipitation, temperature, groundwater withdrawal, GWL during the previous month, and river flow. In total, 11 input patterns from various combinations of variables were developed. About 70% of the data were used to train the models, while the rest were used for validation. In a second step, several metaheuristic algorithms, such as genetic algorithm (GA), particle swarm optimization (PSO), ant colony optimization for continuous domains (ACOR), and differential evolution (DE) were used to improve the model and, consequently, prediction performance. The results showed that (i) RMSE, MAPE, and NSE of 0.51 m, 0.00037 m, and 0.86, respectively, were obtained for the ANFIS model using all input variables, indicating a rather poor performance, (ii) metaheuristic algorithms were able to optimize the parameters of the ANFIS model in predicting GWL, (iii) the input pattern that included all input variables resulted in the most appropriate performance with RMSE, MAPE, and NSE of 0.28 m, 0.00019 m, and 0.97, respectively, using the ANIFS-ACOR hybrid model, (iv) results of Taylor's diagram (CC = 0.98, STD = 0.2, and RMSD = 0.30), as well as the scatterplot (R2 = 0.97), showed that best prediction was achieved by ANFIS-ACOR, and (v) temperature and evaporation exerted stronger influence on GWL prediction than groundwater withdrawal and precipitation. The findings of this study reveal that metaheuristic algorithms can significantly improve the performance of the ANFIS model in predicting GWL. [ABSTRACT FROM AUTHOR]

Successful implementation of construction projects worldwide calls for a set of effective risk management plans in which uncertainties associated with risks and effective response strategies are addressed meticulously. Thus, this study aims to provide an optimization approach with which risk response strategies that maximize the utility function are selected. This selection is by opting for the most appropriate strategies with the highest impact on the project regarding the weight of each risk and budget constraints. Moreover, the risk assessment and response strategy of a construction project in Iran as a case study, based on the global standard of the project management body of knowledge (PMBOK) and related literature, is evaluated. To handle the complexity of the proposed model, different state of the art metaheuristic algorithms including the ant lion optimizer (ALO), dragonfly algorithm (DA), grasshopper optimization algorithm (GOA), Harris hawks optimization (HHO), moth-flame optimization algorithm (MFO), multi-verse optimizer (MVO), sine cosine algorithm (SCA), salp swarm algorithm (SSA), whale optimization algorithm (WOA), and grey wolf optimizer (GWO). These algorithms are validated by the exact solver from CPLEX software and compare with each other. One finding from this comparison is the high performance of MFO and HHO algorithms. Based on some sensitivity analyses, an extensive discussion is provided to suggest managerial insights for real-world construction projects. [ABSTRACT FROM AUTHOR]

Abstract: This paper treats with K-shortest viable path problem in a transportation network including multiple modes. The considered modes are metro, rapid-bus, private and walking. In this network, a viable path is one that the number of mode changes is limited and the traverse time and also the walking, metro and private usage are restricted subjected to some constraints. The traverse time is defined dependent on congestion level. Because constrained shortest path is NP-hard, we extend two metaheuristic algorithms namely GASA and BACS for the given problem. GASA is a Greedy Algorithm Simulated Annealing and BACS is a bi-direction searching Ant Colony System made by two colonies of ants. We evaluate the validation of these algorithms applying several multimodal random networks. In addition, their results are compared with CPLEX 12.1. We find that GASA is appropriate for small networks and BACS has better performance in median and large-scale networks. Our results on Tehran network also demonstrate that BACS provides better objective values than BACS ones because Tehran public transportation is sparse. [Copyright &y& Elsevier]