De novo molecular design is the process of learning knowledge from existing data to propose new chemical structures that satisfy the desired properties. By using de novo design to generate compounds in a directed manner, better solutions can be obtained in large chemical libraries with less comparison cost. But drug design needs to take multiple factors into consideration. For example, in polypharmacology, molecules that activate or inhibit multiple target proteins produce multiple pharmacological activities and are less susceptible to drug resistance. However, most existing molecular generation methods either focus only on affinity for a single target or fail to effectively balance the relationship between multiple targets, resulting in insufficient validity and desirability of the generated molecules. To address the problems, an approach called clustered Pareto-based reinforcement learning (CPRL) is proposed. In CPRL, a pre-trained model is constructed to grasp existing molecular knowledge in a supervised learning manner. In addition, the clustered Pareto optimization algorithm is presented to find the best solution between different objectives. The algorithm first extracts an update set from the sampled molecules through the designed aggregation-based molecular clustering. Then, the final reward is computed by constructing the Pareto frontier ranking of the molecules from the updated set. To explore the vast chemical space, a reinforcement learning agent is designed in CPRL that can be updated under the guidance of the final reward to balance multiple properties. Furthermore, to increase the internal diversity of the molecules, a fixed-parameter exploration model is used for sampling in conjunction with the agent. The experimental results demonstrate that CPRL is capable of balancing multiple properties of the molecule and has higher desirability and validity, reaching 0.9551 and 0.9923, respectively.

Balancing the water consumption of agricultural and ecological is the key point of sustainable social and economic development in an inland river basin. The growth of desert riparian forests in inland river basins mainly depends on a certain phreatic water table depth (PWTD). The main object of this study was to allocate and schedule water resources to regulate the PWTD and satisfy agricultural water demand. Therefore, a multi-objective double layer optimal allocation and scheduling framework based on the computationally efficient integrated surface water-groundwater model (ISGWM), which can simulate the surface water processes, groundwater recharge and discharge processes, and PWTD changes, was constructed and applied to the mainstream of Tarim River Basin (TRB). The top layer model of the framework is an optimal ecological water allocation model, and its optimal allocation results are used as the initial solution of the bottom layer model. The results show that under 5 different inflow frequencies, the agricultural water shortage rate is 0, 17.38 %, 17.41 %, 14.06 %, and 19.94 %, respectively. The PWTD regulation has a great performance. After the optimal scheduling, the proportions of good growth of the control area behind the gate under different inflow frequencies were 98.18 %, 98.18 %, 98.18 %, 90.91 %, and 94.55 %. Agricultural water shortage is mainly due to the non-uniformity distribution of intra-annual inflow and the lack of controlling hydraulic engineering. The regulation of PWTD can guarantee the growth of desert riparian forests on both sides of the mainstream of TRB. Besides, we explored the feasibility of exploiting groundwater to supplement agricultural water consumption. The groundwater exploitation should be controlled within the scope of not causing excessive increase of PWTD (difference between PWTD and target depth <1 m), due to the groundwater exploitation to supplement agricultural water will lead to the increase of PWTD. Overall, this framework, which regulates the PWTD with the change of ecological water supply based on the ISGWM, provides a new idea for the allocation and scheduling of agricultural and ecological water resources in arid inland river basins. It also provides a new method for the coupled cooperative operation of surface water and groundwater.

Poultry managers can better understand the state of poultry through poultry behavior analysis. As one of the key steps in behavior analysis, the accurate estimation of poultry posture is the focus of this research. This study mainly analyzes a top-down pose estimation method of multiple chickens. Therefore, we propose the \"multi-chicken pose\" (MCP), a pose estimation system for multiple chickens through deep learning. Firstly, we find the position of each chicken from the image via the chicken detector; then, an estimate of the pose of each chicken is made using a pose estimation network, which is based on transfer learning. On this basis, the pixel error (PE), root mean square error (RMSE), and image quantity distribution of key points are analyzed according to the improved chicken keypoint similarity (CKS). The experimental results show that the algorithm scores in different evaluation metrics are a mean average precision (mAP) of 0.652, a mean average recall (mAR) of 0.742, a percentage of correct keypoints (PCKs) of 0.789, and an RMSE of 17.30 pixels. To the best of our knowledge, this is the first time that transfer learning has been used for the pose estimation of multiple chickens as objects. The method can provide a new path for future poultry behavior analysis.

In humanitarian aid scenarios, the model of cumulative capacitated vehicle routing problem can be used in vehicle scheduling, aiming at delivering materials to recipients as quickly as possible, thus minimizing their wait time. Traditional approaches focus on this metric, but practical implementations must also consider factors such as driver labor intensity and the capacity for on-site decision-making. To evaluate driver workload, the operation times of relief vehicles are typically used, and multi-objective modeling is employed to facilitate on-site decision-making. This paper introduces a multi-objective cumulative capacitated vehicle routing problem considering operation time (MO-CCVRP-OT). Our model is bi-objective, aiming to minimize both the cumulative wait time of disaster-affected areas and the extra expenditures incurred by the excess operation time of rescue vehicles. Based on the traditional grey wolf optimizer algorithm, this paper proposes a dynamic grey wolf optimizer algorithm with floating 2-opt (DGWO-F2OPT), which combines real number encoding with an equal-division random key and ROV rules for decoding; in addition, a dynamic non-dominated solution set update strategy is introduced. To solve MO-CCVRP-OT efficiently and increase the algorithm\'s convergence speed, a multi-objective improved floating 2-opt (F2OPT) local search strategy is proposed. The utopia optimum solution of DGWO-F2OPT has an average value of two fitness values that is 6.22% lower than that of DGWO-2OPT. DGWO-F2OPT\'s average fitness value in the algorithm comparison trials is 16.49% less than that of NS-2OPT. In the model comparison studies, MO-CCVRP-OT is 18.72% closer to the utopian point in Euclidean distance than CVRP-OT.

OCCUPATIONAL APPLICATIONSIn the context of Industry 5.0, our study advances manufacturing factory layout planning by integrating multi-objective optimization with nature-inspired algorithms and a digital human modeling tool. This approach aims to overcome the limitations of traditional planning methods, which often rely on engineers\' expertise and inputs from various functions in a company, leading to slow processes and risk of human errors. By focusing the multi-objective optimization on three primary targets, our methodology promotes objective and efficient layout planning, simultaneously considering worker well-being and system performance efficiency. Illustrated through a pedal car assembly station layout case, we demonstrate how layout planning can transition into a transparent, cross-disciplinary, and automated activity. This methodology provides multi-objective decision support, showcasing a significant step forward in manufacturing factory layout design practices.
Rationale: Integrating multi-objective optimization in manufacturing layout planning addresses simultaneous considerations of productivity, worker well-being, and space efficiency, moving beyond traditional, expert-reliant methods that often overlook critical design aspects. Leveraging nature-inspired algorithms and a digital human modeling tool, this study advances a holistic, automated design process in line with Industry 5.0. Purpose: This research demonstrates an innovative approach to manufacturing layout optimization that simultaneously considers worker well-being and system performance. Utilizing the Non-dominated Sorting Genetic Algorithm II (NSGA-II) and Particle Swarm Optimization (PSO) alongside a Digital Human Modeling (DHM) tool, the study proposes layouts that equally prioritize ergonomic factors, productivity, and area utilization. Methods: Through a pedal car assembly station case, the study illustrates the transition of layout planning into a transparent, cross-disciplinary, and automated process. This method offers objective decision support, balancing diverse objectives concurrently. Results: The optimization results obtained from the NSGA-II and PSO algorithms represent feasible non-dominated solutions of layout proposals, with the NSGA-II algorithm finding a solution superior in all objectives compared to the expert engineer-designed start solution for the layout. This demonstrates the presented method’s capacity to refine layout planning practices significantly. Conclusions: The study validates the effectiveness of combining multi-objective optimization with digital human modeling in manufacturing layout planning, aligning with Industry 5.0’s emphasis on human-centric processes. It proves that operational efficiency and worker well-being can be simultaneously considered and presents future potential manufacturing design advancements. This approach underscores the necessity of multi-objective consideration for optimal layout achievement, marking a progressive step in meeting modern manufacturing’s complex demands.

Gene selection is a process of selecting discriminative genes from microarray data that helps to diagnose and classify cancer samples effectively. Swarm intelligence evolution-based gene selection algorithms can never circumvent the problem that the population is prone to local optima in the process of gene selection. To tackle this challenge, previous research has focused primarily on two aspects: mitigating premature convergence to local optima and escaping from local optima. In contrast to these strategies, this paper introduces a novel perspective by adopting reverse thinking, where the issue of local optima is seen as an opportunity rather than an obstacle. Building on this foundation, we propose MOMOGS-PCE, a novel gene selection approach that effectively exploits the advantageous characteristics of populations trapped in local optima to uncover global optimal solutions. Specifically, MOMOGS-PCE employs a novel population initialization strategy, which involves the initialization of multiple populations that explore diverse orientations to foster distinct population characteristics. The subsequent step involved the utilization of an enhanced NSGA-II algorithm to amplify the advantageous characteristics exhibited by the population. Finally, a novel exchange strategy is proposed to facilitate the transfer of characteristics between populations that have reached near maturity in evolution, thereby promoting further population evolution and enhancing the search for more optimal gene subsets. The experimental results demonstrated that MOMOGS-PCE exhibited significant advantages in comprehensive indicators compared with six competitive multi-objective gene selection algorithms. It is confirmed that the \"reverse-thinking\" approach not only avoids local optima but also leverages it to uncover superior gene subsets for cancer diagnosis.

In the prediction of protein-ligand affinity, the traditional methods require a large amount of computing resources, and have certain limitations in predicting and simulating the structural changes. Although employing data-driven approaches can yield favorable outcomes in deep learning, it entails a lack of interpretability. Some methods may require additional structural information or domain knowledge to support the interpretation, which may limit their applicability. This paper proposes an affinity variational autoencoder (AffinityVAE) using interaction feature mapping and a variational autoencoder, which consists of a multi-objective model capable of end-to-end affinity prediction and drug discovery. In this study, the limitations of affinity prediction in terms of interpretability are tackled by proposing the concept of a protein-ligand interaction feature map. This increases the diversity and quantity of protein-ligand binding data by designing an adaptive autoencoder of target chemical properties to generate new ligands similar to known ligands and adding them to the original training set. AffinityVAE is then retrained using this extended training set to further validate the protein-ligand binding affinity prediction. Comparisons were conducted between the AffinityVAE and recent methods to demonstrate the high efficiency of the proposed model. The experimental results show that AffinityVAE has very high prediction performance, and it has the potential to enhance the diversity and the amount of protein-ligand binding data, which promotes the drug development.

With the coverage of sensor-rich smart devices (smartphones, iPads, etc.), combined with the need to collect large amounts of data, mobile crowd sensing (MCS) has gradually attracted the attention of academics in recent years. MCS is a new and promising model for mass perception and computational data collection. The main function is to recruit a large group of participants with mobile devices to perform sensing tasks in a given area. Task assignment is an important research topic in MCS systems, which aims to efficiently assign sensing tasks to recruited workers. Previous studies have focused on greedy or heuristic approaches, whereas the MCS task allocation problem is usually an NP-hard optimisation problem due to various resource and quality constraints, and traditional greedy or heuristic approaches usually suffer from performance loss to some extent. In addition, the platform-centric task allocation model usually considers the interests of the platform and ignores the feelings of other participants, to the detriment of the platform\'s development. Therefore, in this paper, deep reinforcement learning methods are used to find more efficient task assignment solutions, and a weighted approach is adopted to optimise multiple objectives. Specifically, we use a double deep Q network (D3QN) based on the dueling architecture to solve the task allocation problem. Since the maximum travel distance of the workers, the reward value, and the random arrival and time sensitivity of the sensing tasks are considered, this is a dynamic task allocation problem under multiple constraints. For dynamic problems, traditional heuristics (eg, pso, genetics) are often difficult to solve from a modeling and practical perspective. Reinforcement learning can obtain sub-optimal or optimal solutions in a limited time by means of sequential decision-making. Finally, we compare the proposed D3QN-based solution with the standard baseline solution, and experiments show that it outperforms the baseline solution in terms of platform profit, task completion rate, etc., the utility and attractiveness of the platform are enhanced.

Data gathering in wireless sensor networks (WSNs) is vital for deploying and enabling WSNs with the Internet of Things (IoTs). In various applications, the network is deployed in a large-scale area, which affects the efficiency of the data collection, and the network is subject to multiple attacks that impact the reliability of the collected data. Hence, data collection should consider trust in sources and routing nodes. This makes trust an additional optimization objective of the data gathering in addition to energy consumption, traveling time, and cost. Joint optimization of the goals requires conducting multiobjective optimization. This article proposes a modified social class multiobjective particle swarm optimization (SC-MOPSO) method. The modified SC-MOPSO method is featured by application-dependent operators named interclass operators. In addition, it includes solution generation, adding and deleting rendezvous points, and moving to the upper and lower class. Considering that SC-MOPSO provides a set of nondominated solutions as a Pareto front, we employed one of the multicriteria decision-making (MCDM) methods, i.e., simple additive sum (SAW), for selecting one of the solutions from the Pareto front. The results show that both SC-MOPSO and SAW are superior in terms of domination. The set coverage of SC-MOPSO is 0.06 dominant over NSGA-II compared with only a mastery of 0.04 of NSGA-II over SC-MOPSO. At the same time, it showed competitive performance with NSGA-III.

A novel multi-objective Coronavirus disease optimization algorithm (MOCOVIDOA) is presented to solve global optimization problems with up to three objective functions. This algorithm used an archive to store non-dominated POSs during the optimization process. Then, a roulette wheel selection mechanism selects the effective archived solutions by simulating the frameshifting technique Coronavirus particles use for replication. We evaluated the efficiency by solving twenty-seven multi-objective (21 benchmarks & 6 real-world engineering design) problems, where the results are compared against five common multi-objective metaheuristics. The comparison uses six evaluation metrics, including IGD, GD, MS, SP, HV, and delta p (ΔP). The obtained results and the Wilcoxon rank-sum test show the superiority of this novel algorithm over the existing algorithms and reveal its applicability in solving multi-objective problems.