Abstract This study explores crop yield forecasting through weight agnostic neural networks (WANN) optimized by a modified metaheuristic. WANNs offer the potential for lighter networks with shared weights, utilizing a two-layer cooperative framework to optimize network architecture and shared weights. The proposed metaheuristic is tested on real-world crop datasets and benchmarked against state-of-the-art algorithms using standard regression metrics. While not claiming WANN as the definitive solution, the model demonstrates significant potential in crop forecasting with lightweight architectures. The optimized WANN models achieve a mean absolute error (MAE) of 0.017698 and an R-squared ( $$R^2$$ R 2 ) score of 0.886555, indicating promising forecasting performance. Statistical analysis and Simulator for Autonomy and Generality Evaluation (SAGE) validate the improvement significance and feature importance of the proposed approach.

Water quality prediction is an indispensable task in water environment and source management. The existing predictive models are mainly designed by data-driven artificial neural networks (ANNs), especially deep learning models for large-scale water quality prediction. However, the state of water environment is a dynamic process where the stationarity of water quality data suffers from time variation and human activities, which leads to a poor prediction accuracy because ANNs receive whole water quality data passively, including abnormal conditions. We consider such a tough problem in this article and propose an event-triggered deep fuzzy neural network (ET-DFNN) to pursue the better performance of water quality prediction in the complex water environment. First, a deep pretraining model is constructed to extract the effective features from raw water quality data. Second, we construct a DFNN model where the extracted effective features are considered as the input variables. Third, some events are defined to characterize the abnormal conditions of state evolution in water quality. The DFNN is trained and updated using different learning strategies only when the corresponding events are triggered, otherwise it ignores the current data sample and directly goes to the next data sample. The practical data-based experimental results show that the ET-DFNN achieves better prediction performance in accuracy and efficiency than its peers. Especially, the training efficiency of ET-DFNN is improved by 57.94% on total phosphorus prediction and 48.31% on biochemical oxygen demand prediction, respectively.

Assigning shipping orders to the most appropriate driver in the shortest time but with the highest profit is one of the major concerns of transportation companies. Many studies have been conducted on transportation service procurement systems; however, due to the lack of a framework for modeling human interactions, none of them has used the concept of information diffusion for this purpose. In this article, a monoplex weighted drivers’ collaboration network is developed to model drivers’ relationships within the transportation system. Besides, to identify and track communities during a given time interval of the network, a new community detection algorithm, called dynamic overlapping community detection (DOCD) algorithm, is designed, which can identify viable leaders in each community. In addition to detecting community leaders, the algorithm is able to monitor, assess, and detect the durability of these community leaders over time, which other algorithms are not able to. To evaluate the performance of the algorithm, it is compared with five different algorithms in terms of 14 evaluation measures. The results show the proposed DOCD algorithm outperforms the other algorithms with 88% superiority in the evaluation measures. Then, a fuzzy profit maximization model is developed using information diffused by the identified communities’ viable leaders and information diffusion power of each community. Analyzing a real case study obtains two achievements in the form of “high-risk scenario” and “low-risk scenario” for well-known and novice transportation companies, respectively. Therefore, the obtained results show that transportation companies allocate orders to drivers based on their reputation and risk levels.

In today's competitive environment, it is essential to design a flexible-responsive manufacturing system with automatic material handling systems. In this article, a fuzzy mixed integer linear programming model is designed for cell formation problems including the scheduling of parts within cells in a cellular manufacturing system (CMS) where several automated guided vehicles (AGVs) are in charge of transferring the exceptional parts. Notably, using these AGVs in CMS can be challenging from the perspective of mathematical modeling due to consideration of AGVs’ collision as well as parts pickup/delivery. This article investigates the role of AGVs and human factors as indispensable components of automation systems in the cell formation and scheduling of parts under fuzzy processing time. The proposed objective function includes minimizing the makespan and intercellular movements of parts. Due to the NP-hardness of the problem, a hybrid genetic algorithm (GA/heuristic) and a whale optimization algorithm (WOA) are developed. The experimental results reveal that our proposed algorithms have a high performance compared to CPLEX and the other two well-known algorithms, i.e., particle swarm optimization and ant colony optimization, in terms of computational efficiency and accuracy. Finally, WOA stands out as the best algorithm to solve the problem.

Abstract Nowadays, transportation and logistics are considered as the drivers of economic development in the countries due to their impacts on the main variables of the country's economy such as production, employment, price, and the cost of living. Statistics indicate that fuel consumption constructs a major part of transportation costs, where its optimization leads to the creation of an energy‐efficient and sustainable transportation system. On the other hand, vehicles' traffic is also one of the main criteria affecting the travel time of vehicles between demand nodes in a supply chain, increasing fuel consumption, and, consequently, damaging effects of greenhouse gasses. In this paper, a novel robust mixed‐integer linear programming model is developed for a green vehicle routing problem with intermediate depots considering different urban traffic conditions, fuel consumption, time windows of services, and uncertain demand for perishable products. To validate and solve the suggested model, CPLEX solver of GAMS software is employed as an exact method. Finally, a case study problem is investigated to evaluate the applicability of the proposed model and determine the optimal managerial insights and policies in the real‐world conditions using sensitivity analyses. Moreover, a novel robustness threshold comparison is conducted to find the optimal level of budget assignment.

Flow shop scheduling (FSS) problem constitutes a major part of production planning in every manufacturing organization. It aims at determining the optimal sequence of processing jobs on available machines within a given customer order. In this article, a novel biobjective mixed-integer linear programming (MILP) model is proposed for FSS with an outsourcing option and just-in-time delivery in order to simultaneously minimize the total cost of the production system and total energy consumption. Each job is considered to be either scheduled in-house or to be outsourced to one of the possible subcontractors. To efficiently solve the problem, a hybrid technique is proposed based on an interactive fuzzy solution technique and a self-adaptive artificial fish swarm algorithm (SAAFSA). The proposed model is treated as a single objective MILP using a multiobjective fuzzy mathematical programming technique based on the ε-constraint, and SAAFSA is then applied to provide Pareto optimal solutions. The obtained results demonstrate the usefulness of the suggested methodology and high efficiency of the algorithm in comparison with CPLEX solver in different problem instances. Finally, a sensitivity analysis is implemented on the main parameters to study the behavior of the objectives according to the real-world conditions.