Product demand forecasting plays a vital role in supply chain management since it is directly related to the profit of the company. According to companies’ concerns regarding product demand forecasting, many researchers have developed various forecasting models in order to improve accuracy. We propose a hybrid forecasting model called GA-GRU, which combines Genetic Algorithm (GA) with Gated Recurrent Unit (GRU). Because many hyperparameters of GRU affect its performance, we utilize GA that finds five kinds of hyperparameters of GRU including window size, number of neurons in the hidden state, batch size, epoch size, and initial learning rate. To validate the effectiveness of GA-GRU, this paper includes three experiments: comparing GA-GRU with other forecasting models, k-fold cross-validation, and sensitive analysis of the GA parameters. During each experiment, we use root mean square error and mean absolute error for calculating the accuracy of the forecasting models. The result shows that GA-GRU obtains better percent deviations than other forecasting models, suggesting setting the mutation factor of 0.015 and the crossover probability of 0.70. In short, we observe that GA-GRU can optimally set five types of hyperparameters and obtain the highest forecasting accuracy.

In energy supply chain management (ESCM), the supply chain members try to make long-term contracts for supplying energy stably and reducing the cost. Currently, optimizing ESCM is a complex problem with two social issues: environmental regulations and uncertainties. First, environmental regulations have been tightened in countries around the world, leading to eco-friendly management. As a result, it has become imperative for the energy buyer to consider not only the total operating cost but also carbon emissions. Second, the uncertainties, such as pandemics and wars, have had a serious impact on handling ESCM. Since the COVID-19 pandemic disrupted the supply chain, the supply chain members adopted emergency procurement for sustainable operations. In this study, we developed an optimization model using mixed-integer linear programming to solve ESCM with supplier selection problems in emergency procurement. The model considers a single thermal power plant and multiple fossil fuel suppliers. Because of uncertainties, energy demand may suddenly change or may not be supplied on time. To better manage these uncertainties, we developed a rolling horizon method (RHM), which is a well-known method for solving deterministic problems in mathematical programming models. To test the model and the RHM, we conducted three types of numerical experiments. First, we examined replenishment strategies and schedules under uncertain demands. Second, we conducted a supplier selection experiment within a limited budget and carbon emission regulations. Finally, we conducted a sensitivity analysis of carbon emission limits. The results show that our RHM can handle ESCM under uncertain situations effectively.

The economic development of a country is profoundly influenced by how it protects and utilizes its natural resources, one of which is wastewater, with emphasis on environmental sustainability. For this multi-factorial sustainability analysis, this study introduces a novel Fuzzy Parameterized Fuzzy Hypersoft Set (FPFHSS) structure to develop a comprehensive decision-making and performance evaluation system for wastewater treatment facilities. The practicality of the designed system is explored by highlighting its ability to evaluate urban projects and work as a decision-making system for environmental policy design. The study introduces an algorithm based on the hypersoft structure, allowing the division of attributes into sub-attributes for a more concise analysis. The sub-parametric values are first parameterized into fuzzy numbers and then evaluated based on their relative importance. With proper fuzzy parameterization of each sub-attribute, novel specialized FPFHSS based Technique of Ordered Preference Similar to Ideal Solution (TOPSIS) and Multi-Objective Optimization on the basis of a Ratio Analysis (MULTIMOORA) approaches are developed that provide a versatile analysis in addition to the newly developed algorithm. With these 3 specialized systems, 4 case studies were developed based on 19 pseudo-realistic environmental, social, technical, and economic factors each simulating a different scenario faced by nations highlighting sustainability, technical performance and the environment allowing for informed decision-making while addressing uncertainty making them highly suitable as a computational AI solution for real-data analysis. This fuzzy analysis offers a reference for making informed decisions in the context of environmental remediation and complex scenario simulations.

Recently, as global warming has become a major issue, many companies have increased their efforts to control carbon emissions in green supply chain management (GSCM) activities. This paper deals with the multi-item replenishment problem in GSCM, from both economic and environmental perspectives. A single buyer orders multiple items from a single supplier, and simultaneously considers carbon cap-and-trade under limited storage capacity and limited budget. In this case we can apply a can-order policy, which is a well-known multi-item replenishment policy. Depending on the market characteristics, we develop two mixed-integer programming (MIP) models based on the can-order policy. The deterministic model considers a monopoly market in which a company fully knows the market information, such that both storage capacity and budget are already determined. In contrast, the fuzzy model considers a competitive or a new market, in which case both of those resources are considered as fuzzy numbers. We performed numerical experiments to validate and assess the efficiency of the developed models. The results of the experiments showed that the proposed can-order policy performed far better than the traditional can-order policy in GSCM. In addition, we verified that the fuzzy model can cope with uncertainties better than the deterministic model in terms of total expected costs.

Flow shop scheduling has garnered significant attention from researchers over the past ten years, establishing itself as a prominent area of study within the field of scheduling. Nevertheless, there exists a paucity of research dedicated to addressing Non-Permutation Flow Shop Scheduling Problems. In this study, a Hybrid Evolution Strategies (HES) is suggested by combining the exploitation ability of Nawaz, Enscore, and Ham (NEH) Heuristic, the exploration ability of Improved Evolution Strategies (IES), and a Local Search Technique to minimize the makespan of NPFSSP. The primary solution is produced through the NEH Heuristic, serving as a foundational solution for the IES. The IES is applied in two stages, in the first stage it improves the permutation sequence found from the NEH heuristic. In the second stage of the IES, the permutation sequence on the first 40% of machines is fixed as found in the first stage. The sequence on the last 60% of machines is altered only so that the makespan is minimized and a good non-permutation sequence is found. Recombination and mutation are the main genetic operators in IES. For recombination in IES, 16 offspring are generated randomly from a single parent. The Quad swap mutation operator is employed in the IES to optimize the utilization of the solution space while minimizing computational time. To prevent trapping in local minima, a Local Search Technique is integrated into the IES algorithm, which guides solutions to less explored areas. Computational analyses indicate that HES exhibits superior performance regarding solution quality, computational efficiency, and robustness.