The crayfish optimization algorithm (COA), proposed in 2023, is a metaheuristic optimization algorithm that is based on crayfish’s summer escape behavior, competitive behavior, and foraging behavior. COA has a good optimization performance, but it still suffers from the problems of slow convergence speed and sensitivity to the local optimum. To solve these problems, an improved multi-strategy crayfish optimization algorithm for solving numerical optimization problems, called IMCOA, is proposed to address the shortcomings of the original crayfish optimization algorithm for each behavioral strategy. Aiming at the imbalance between local exploitation and global exploration in the summer heat avoidance and competition phases, this paper proposes a cave candidacy strategy and a fitness–distance balanced competition strategy, respectively, so that these two behaviors can better coordinate the global and local optimization capabilities and escape from falling into the local optimum prematurely. The directly foraging formula is modified during the foraging phase. The food covariance learning strategy is utilized to enhance the population diversity and improve the convergence accuracy and convergence speed. Finally, the introduction of an optimal non-monopoly search strategy to perturb the optimal solution for updates improves the algorithm’s ability to obtain a global best solution. We evaluated the effectiveness of IMCOA using the CEC2017 and CEC2022 test suites and compared it with eight algorithms. Experiments were conducted using different dimensions of CEC2017 and CEC2022 by performing numerical analyses, convergence analyses, stability analyses, Wilcoxon rank–sum tests and Friedman tests. Experiments on the CEC2017 and CEC2022 test suites show that IMCOA can strike a good balance between exploration and exploitation and outperforms the traditional COA and other optimization algorithms in terms of its convergence speed, optimization accuracy, and ability to avoid premature convergence. Statistical analysis shows that there is a significant difference between the performance of the IMCOA algorithm and other algorithms. Additionally, three engineering design optimization problems confirm the practicality of IMCOA and its potential to solve real-world problems.

Purpose: The purpose of this paper is to explore the operation strategies of a manufacturer who produces brown and green product simultaneously.Design/methodology/approach: The authors establish three models to examine the joint decisions of pricing and advertising. Three advertising strategies are: non-advertising investment (NA), advertising investment for brown product (BA) and advertising investment for green product (GA).Findings: The theoretical analysis shows that advertising investment can substantially increase the product greening level and manufacturer's profit. More importantly, we find that the GA strategy is more likely to be the best strategy as the advertising investment efficiency increases. The BA strategy is more likely to be preferred as the R&D cost increases. Finally, the modeling results are verified by numerical experiments, and more insights are obtained.Research limitations/implications: This paper considers the case in which a single manufacturer produces the brown and green product simultaneously. In fact, many manufacturers in the market produce brown and green product at the same time. Furthermore, in addition to advertising investment for brown product and green product, manufacturers can also invest in advertising for brands.Originality/value: The paper contributes to the investigations on green production and advertising decisions of a manufacturer who produces brown and green products simultaneously.

With the rapid development of Online to Offline (O2O) business, millions of transactions each day along with the varying processing time of merchants and the complexity of traffic conditions pose significant challenges to effective and efficient delivery of orders. This paper studies the complex adaptive dynamics of O2O platforms by combining the behaviors of customers, merchants, dispatcher, and couriers in the context of a multi-agent model. Serving as a testbed, the simulation model enables the evaluation of alternative order delivery strategies. Preliminary experimental results show that TSP-based delivery strategy is more efficient than the nearer merchant assignment strategy. As an important property, the larger load capacity of couriers is beneficial to improve the completion rate of orders rather than the completion time. Finally, the experiment using the real road network and the real order data demonstrates the applicability of the proposed multi-agent model of O2O platform in the real scenario.

Many real-world applications often generate unbalanced data. Learning from such data may lead to biased classifiers that perform poorly on the class of interest. Oversampling methods have been shown to be effective in re-balancing unbalanced data to help classifiers avoid performance bias. However, many existing oversampling methods rely on a pre-designed linear model structure and the neighborhood information of an original instance. This may lead to the generation of noisy instances when the original data has noise. In this study, we develop a novel oversampling method in which genetic programming is introduced to automatically select good-quality instances and evolve a model structure that combines the selected instances to create a new instance. In the proposed oversampling method, an individual is used to represent a generated instance, which is evaluated by the fitness function designed based on the Euclidean distance and cosine theorem. In the experiments, we examine the effectiveness of the proposed oversampling method in assisting different types of classifiers to solve the issue of class imbalance, and compare it with popular sampling methods in unbalanced classification. The results have been analyzed comprehensively, indicating that the new method successfully addressed the class imbalance issue by generating a group of good-quality instances for the minority class and outperformed the compared sampling methods in almost all cases.