Simulated Annealing Mechanism Research Articles

Hybrid Aquila optimizer for efficient classification with probabilistic neural networks

The model of a probabilistic neural network (PNN) is commonly utilized for classification and pattern recognition issues in data mining. An approach frequently used to enhance its effectiveness is the adjustment of PNN classifier parameters through the outcomes of metaheuristic optimization strategies. Since PNN employs a limited set of instructions, metaheuristic algorithms provide an efficient way to modify its parameters. In this study, we have employed the Aquila optimizer algorithm (AO), a contemporary algorithm, to modify PNN parameters. We have proposed two methods: Aquila optimizer based probabilistic neural network (AO-PNN), which uses both local and global search capabilities of AO, and hybrid Aquila optimizer and simulated annealing based probabilistic neural network (AOS-PNN), which integrates the global search abilities of AO with the local search mechanism of simulated annealing (SA). Our experimental results indicate that both AO-PNN and AOS-PNN perform better than the PNN model in terms of accuracy across all datasets. This suggests that they have the potential to generate more precise results when utilized to improve PNN parameters. Moreover, our hybridization technique, AOS-PNN, is more effective than AO-PNN, as evidenced by classification experiments accuracy, data distribution, convergence speed, and significance. We have also compared our suggested approaches with three different methodologies, namely Coronavirus herd immunity optimizer based probabilistic neural network (CHIO-PNN), African buffalo algorithm based probabilistic neural network (ABO-PNN), and β-hill climbing. We have found that AO-PNN and AOS-PNN have achieved significantly higher classification accuracy rates of 90.68 and 93.95, respectively.

An enhanced dynamic differential annealed algorithm for global optimization and feature selection

Abstract Dynamic differential annealed optimization (DDAO) is a recently developed physics-based metaheuristic technique that mimics the classical simulated annealing mechanism. However, DDAO has limited search abilities, especially when solving complicated and complex problems. A unique variation of DDAO, dubbed as mDDAO, is developed in this study, in which opposition-based learning technique and a novel updating equation are combined with DDAO. mDDAO is tested on 10 different functions from CEC2020 and compared with the original DDAO and nine other algorithms. The proposed mDDAO algorithm performance is evaluated using 10 numerical constrained functions from the recently released CEC 2020 benchmark suite, which includes a variety of dimensionally challenging optimisation tasks. Furthermore, to measure its viability, mDDAO is employed to solve feature selection problems using fourteen UCI datasets and a real-life Lymphoma diagnosis problem. Results prove that mDDAO has a superior performance and consistently outperforms counterparts across benchmarks, achieving fitness improvements ranging from 1% to 99.99%. In feature selection, mDDAO excels by reducing feature count by 23% to 79% compared to other methods, enhancing computational efficiency and maintaining classification accuracy. Moreover, in lymphoma diagnosis, mDDAO demonstrates up to 54% higher average fitness, 18% accuracy improvement, and 86% faster computation times.

Vehicle Routing Problem with Drones Considering Time Windows and Dynamic Demand

As a new delivery mode, the collaborative delivery of packages using trucks and drones has been proven to reduce delivery costs and delivery time. To cope with the huge cost challenges brought by strict time constraints and ever-changing customer orders in the actual delivery process, we established a two-stage optimization model based on different demand response strategies with the goal of minimizing delivery costs. To solve this problem, we designed a simulated annealing chimp optimization algorithm with a sine–cosine operator. The performance of this algorithm is improved by designing a variable-dimensional matrix encode to generate an initial solution, incorporating a sine–cosine operator and a simulated annealing mechanism to avoid falling into a local optimum. Numerical experiments verify the effectiveness of the proposed algorithm and strategy. Finally, we analyze the impact of dynamic degree on delivery cost. The proposed model and algorithm extend the theory of the vehicle routing problem with drones and also provide a feasible solution for route planning, taking into account dynamic demands and time windows.

An enhanced decision-making framework for predicting future trends of sharing economy.

This work aims to provide a reliable and intelligent prediction model for future trends in sharing economy. Moreover, it presents valuable insights for decision-making and policy development by relevant governmental bodies. Furthermore, the study introduces a predictive system that incorporates an enhanced Harris Hawk Optimization (HHO) algorithm and a K-Nearest Neighbor (KNN) forecasting framework. The method utilizes an improved simulated annealing mechanism and a Gaussian bare bone structure to improve the original HHO, termed SGHHO. To achieve optimal prediction performance and identify essential features, a refined simulated annealing mechanism is employed to mitigate the susceptibility of the original HHO algorithm to local optima. The algorithm employs a mechanism that boosts its global search ability by generating fresh solution sets at a specific likelihood. This mechanism dynamically adjusts the equilibrium between the exploration and exploitation phases, incorporating the Gaussian bare bone strategy. The best classification model (SGHHO-KNN) is developed to mine the key features with the improvement of both strategies. To assess the exceptional efficacy of the SGHHO algorithm, this investigation conducted a series of comparative trials employing the function set of IEEE CEC 2014. The outcomes of these experiments unequivocally demonstrate that the SGHHO algorithm outperforms the original HHO algorithm on 96.7% of the functions, substantiating its remarkable superiority. The algorithm can achieve the optimal value of the function on 67% of the tested functions and significantly outperforms other competing algorithms. In addition, the key features selected by the SGHHO-KNN model in the prediction experiment, including " Form of sharing economy in your region " and " Attitudes to the sharing economy ", are important for predicting the future trends of the sharing economy in this study. The results of the prediction demonstrate that the proposed model achieves an accuracy rate of 99.70% and a specificity rate of 99.38%. Consequently, the SGHHO-KNN model holds great potential as a reliable tool for forecasting the forthcoming trajectory of the sharing economy.

Multi-objective optimization of energy-efficient remanufacturing system scheduling problem with lot-streaming production mode

Most previous studies on the scheduling problem in remanufacturing systems have focused on single or two production stages and economic criteria such as makespan or tardiness. However, environmental criteria like energy consumption or carbon emissions, as well as the lot-streaming production mode, have not been adequately addressed. This paper presents a novel approach to the energy-efficient remanufacturing system scheduling problem with lot-streaming production mode (ERSSP-LS) for the first time, which integrates disassembly, reprocessing, and reassembly three pivotal production stages. Firstly, a multi-objective mathematical model aimed at simultaneously minimizing the total energy consumption (TEC) and makespan (Cmax) is established and presented. To further reduce the TEC, the well-accepted energy-saving measure, known as the turn off and on strategy, is also integrated. Subsequently, a hybrid multi-objective optimization algorithm called HMOFFO, which combines the fruit fly optimization algorithm and simulated annealing mechanism is developed to seek the promising Pareto solution set. Finally, several experiments on a real-life case and a group of ten random test instances are carried out and the computational results affirm the feasibility and effectiveness of the proposed HMOFFO in solving the ERSSP-LS.

Adaptive large neighborhood search algorithm for multi-stage weapon target assignment problem

Weapon target assignment (WTA) is a critical operational research topic that can be applied to plentiful military fields. A new variant of WTA, the multi-stage WTA problem, is investigated, which is to assign limited weapons to all targets in multiple attacking phases. The attacking flexibility for targets in different stages is considered, and a binary nonlinear integer programming model is developed to formulate the problem. An improved adaptive large-scale neighborhood search (ALNS) algorithm is proposed. First, a priority-based encoding strategy is designed to facilitate the feasible solution generating and solution space exploring. Then, six specific neighborhood structures are designed to generate 15 operators through their combination. The simulated annealing mechanism is integrated to avoid getting trapped in local optima. Moreover, an adaptive learning strategy is employed to improve the exploration capability. Both exact methods and metaheuristics are compared with the ALNS algorithm by instances with different scales. Experimental results show that, for most situations, the ALNS algorithm can obtain higher-quality solutions in a shorter time.

Multi-Terminal Berth and Quay Crane Joint Scheduling in Container Ports Considering Carbon Cost

As container ports become increasingly important to the global supply chain, a growing number of ports are improving their competencies by consolidating multiple terminal resources internally. In addition, in the context of energy conservation and emission reduction, ports measure competitiveness not only in terms of terminal size, throughput and service level, but also in terms of low energy consumption and low pollution. Therefore, a nonlinear mixed-integer programming model considering the cost of carbon is developed for the multi-terminal berth and quay crane joint robust scheduling problem under uncertain environments to minimize the sum of expectation and variance of total cost under all randomly generated samples. The model considers the water depth and interference of quay cranes, etc. The expected vessel arrival time and the average operational efficiency plus relaxation are used as their actual values when scheduling. Finally, an improved adaptive genetic algorithm is developed by combining the simulated annealing mechanism, and numerical experiments are designed. The results show that the joint berth and quay crane scheduling with uncertainties and a multi-terminal coordination mechanism can effectively reduce the operating cost, including carbon costs and the vessel departure delay rate, and can improve resource utilization. Meanwhile, the scheduling with the multi-terminal coordination mechanism can obtain more significant improvement effects than the scheduling with uncertainties.

Local Path Planning with Multiple Constraints for USV Based on Improved Bacterial Foraging Optimization Algorithm

The quality of unmanned surface vehicle (USV) local path planning directly affects its safety and autonomy performance. The USV local path planning might easily be trapped into local optima. The swarm intelligence optimization algorithm is a novel and effective method to solve the path-planning problem. Aiming to address this problem, a hybrid bacterial foraging optimization algorithm with a simulated annealing mechanism is proposed. The proposed algorithm preserves a three-layer nested structure, and a simulated annealing mechanism is incorporated into the outermost nested dispersal operator. The proposed algorithm can effectively escape the local optima. Convention on the International Regulations for Preventing Collisions at Sea (COLREGs) rules and dynamic obstacles are considered as the constraints for the proposed algorithm to design different obstacle avoidance strategies for USVs. The coastal port is selected as the working environment of the USV in the visual test platform. The experimental results show the USV can successfully avoid the various obstacles in the coastal port, and efficiently plan collision-free paths.

The Performance of Electronic Current Transformer Fault Diagnosis Model: Using an Improved Whale Optimization Algorithm and RBF Neural Network

With the widely application of electronic transformers in smart grids, transformer faults have become a pressing problem. However, reliable fault diagnosis of electronic current transformers (ECT) is still an open problem due to the complexity and diversity of fault types. In order to solve this problem, this paper proposes an ECT fault diagnosis model based on radial basis function neural network (RBFNN) and optimizes the model parameters and the network size of RBFNN simultaneously via an improved whale optimization algorithm (WOA) to improve the classification accuracy and robustness of RBFNN. Since the classical WOA is easy to fall into a locally optimal performance, a hybrid multi-strategies WOA algorithm (CASAWOA) is proposed for further improvement in optimization performance. Firstly, we introduced the tent chaotic map strategy to improve the population diversity of WOA. Secondly, we introduced nonlinear convergence factor and adaptive inertia weight to enhance the exploitation ability of the WOA. Finally, on the premise of ensuring the convergence speed of the algorithm, we modified the simulated annealing mechanism in order to prevent premature convergence. The benchmark function tests show that the CASAWOA outperforms other state-of-the-art WOA algorithms in terms of convergence speed and exploration ability. Furthermore, to validate the performance of ECT fault diagnosis model based on CASAWOA-RBFNN, a comprehensive analysis of eight fault diagnosis methods is conducted based on the ECT fault samples collected from the detection circuit. The experimental results show that the CASAWOA-RBFNN achieves an accuracy of 97.77% in ECT fault diagnosis, which is 9.8% better than WOA-RBF and which shows promising engineering practicality.

A Multi-strategy Improved Outpost and Differential Evolution Mutation Marine Predators Algorithm for Global Optimization.

Marine Predators Algorithm (MPA) is a recent efficient metaheuristic algorithm that is enlightened by the biological behavior of ocean predators and prey. This algorithm simulates the Levy and Brownian movements of prevalent foraging strategy and has been applied to many complex optimization problems. However, the algorithm has defects such as a low diversity of the solutions, ease into the local optimal solutions, and decreasing convergence speed in dealing with complex problems. A modified version of this algorithm called ODMPA is proposed based on the tent map, the outpost mechanism, and the differential evolution mutation with simulated annealing (DE-SA) mechanism. The tent map and DE-SA mechanism are added to enhance the exploration capability of MPA by increasing the diversity of the search agents, and the outpost mechanism is mainly used to improve the convergence speed of MPA. To validate the outstanding performance of the ODMPA, a series of global optimization problems are selected as the test sets, including the standard IEEE CEC2014 benchmark functions, which are the authoritative test set, three well-known engineering problems, and photovoltaic model parameters tasks. Compared with some famous algorithms, the results reveal that ODMPA has achieved better performance than its counterparts in CEC2014 benchmark functions. And in solving real-world optimization problems, ODMPA could get higher accuracy than other metaheuristic algorithms. These practical results demonstrate that the mechanisms introduced positively affect the original MPA, and the proposed ODMPA can be a widely effective tool in tackling many optimization problems.

An improved algorithm for resolving overlapping peaks in ion mobility spectrometry and its application to the separation of glycan isomers.

Despite the popularity of ion mobility spectrometry (IMS) for glycan analysis, its limited structural resolution hinders the effective separation of many glycan isomers. This leads to the overlap of IMS peaks, consequently impacting the accurate identification of glycan compositions. To this end, an improved algorithm, namely second-order differentiation combined with a simulated annealing particle swarm optimization algorithm based on sine adaptive weights (DWSA-PSO), was proposed for the separation of overlapping IMS peaks formed by glycan isomers. DWSA-PSO first performed second-order differentiation to automatically determine the number of components in overlapping peaks and exclude impossible single-peak combinations. It then introduced sinusoidal adaptive weights and a simulated annealing mechanism to improve the algorithm's search capability and global optimization performance, thereby enabling accurate and efficient separation of individual peaks. To evaluate the performance of DWSA-PSO and its application to the separation of glycan isomers, multiple sets of overlapping peaks with different degrees of overlap were simulated, and various types of multi-component overlapping peaks were formed using six disaccharide and four trisaccharide isomers. The experimental results consistently demonstrated that the DWSA-PSO algorithm outperformed both the improved particle swarm optimization (IPSO) algorithm and the dynamic inertia weight particle swarm optimization (DIWPSO) algorithm in terms of separation accuracy, running time, and fitness values. In addition, the DWSA-PSO algorithm was successfully applied to the separation of glycan isomers in malt milk beverage. All these results reveal the capability of the DWSA-PSO algorithm to facilitate the accurate identification of glycan isomers.

Parameter optimization of moving horizon estimation based on hybrid intelligent algorithm

AbstractIn the application of moving horizon estimation (MHE) algorithm, the window length will affect the estimation accuracy and the computing efficiency. For this kind of problem, a method of parameter optimization is proposed to obtain suitable window length. Firstly, in order to facilitate online solution, the optimization problem involved in the algorithm is transformed into a quadratic programming (QP) problem in matrix form. Secondly, for the time index and the estimated residual index that measure different properties, the normalization idea is adopted to incorporate them into the same dimension to design the fitness function, and a genetic optimization algorithm based on simulated annealing mechanism is given to search for the optimal window length. Finally, the proposed parameter optimization method is verified by two cases. The results show that the parameter optimization method has the advantages of excellent local search ability and sufficient convergence, and the window length obtained by this method can better take into account the two performance indexes of the MHE algorithm and improve the estimation performance.

Optimization and Application of Communication Resource Allocation Algorithm for Urban Rail Transit Planning.

The construction and operation of China's rail transit system have entered a high-speed development stage, and the rapid increase of train speed and mileage has brought greater challenges to the safety and reliability of the rail transit system. Network planning evaluation is the key to the early decision-making of urban rail transit project, which directly determines the success or failure of the whole project. How to scientifically and reasonably evaluate the urban rail transit information resource network planning has become a difficult problem for many urban planners to solve. Therefore, this paper studies the optimization of the communication resource allocation algorithm and the comprehensive evaluation of its application for urban rail transit planning. In this paper, based on CVNN structure, the network prototype is an extension of RVNN structure. In the abstract, its processing unit is composed of a pair of real-number processors that can realize certain operations. HNN is a fully connected recurrent neural network based on the idea of the energy function, which is helpful to understand the calculation mode of HNN, and the research shows that HNN can solve many combinatorial optimization problems. In addition, the combination of neural network and genetic algorithm with simulated annealing mechanism can also bring new directions for research. On the basis of experimental analysis, it can be concluded that in general, the error reduction rate of the optimization scheme designed in this paper can reach 58.6% on average. In practical application, the accuracy of the optimal bit error rate is 52.4%.

Simulated annealing-based dynamic step shuffled frog leaping algorithm: Optimal performance design and feature selection

The shuffled frog leaping algorithm is a new optimization algorithm proposed to solve the combinatorial optimization problem, which effectively combines the memetic algorithm based on a memetic algorithm and the particle swarm algorithm based on population behavior. The algorithm is widely used because it is easy to implement and requires few parameters to be adjusted. However, there are still some characteristics of this method that need to be improved because it is easy to fall into local optimization or poor search ability. To alleviate this limitation, a new version of the improved SFLA is proposed in this paper, which incorporates a dynamic step size adjustment strategy based on historical information, a specular reflection learning mechanism, and a simulated annealing mechanism based on chaotic mapping and levy flight. Firstly, the dynamic step size adjustment strategy based on historical information effectively helps to balance local exploration and global exploitation and alleviates the problem of falling into local optimum. Second, the specular reflection learning mechanism increases the possibility of searching for valid solutions in feasible domains and enhances the search ability of individuals in the population. Finally, an improved simulated annealing strategy is executed for each memetic, which improves the efficiency of local exploitation. In order to test the performance of the proposed algorithm, 31 test functions were selected from IEEE CEC2014 and 23 essential benchmark functions, and comparative experiments were carried out from the two dimensions of 30 and 100. Furthermore, a series of competing algorithms are selected, which involve nine classical standard algorithms, including PSO, BA, SSA, FA, SCA, WOA, GWO, MFO, and SFLA, as well as, six well-known improved algorithms, including LSFLA, DDSFLA, GOTLBO, ALCPSO, BLPSO, CLPSO. Furthermore, Wilcoxon signed-rank test and Friedman test are used as testing tools to illustrate the scalability of the proposed algorithm. From the analysis of the results, it can be seen that this proposed method has been effectively improved concerning stability and the quality of the optimal solution obtained from the search, both in low and high dimensions, and the ability to jump out of the local optimum has also been enhanced. In addition, to prove that this method has a reliable performance in discrete problems and continuous problems, DSSRLFLA is mapped into a discrete space, and 24 UCI data sets are also selected to evaluate the performance of the new feature selection method. The experimental results illustrate that this improved method can obtain fewer features and higher classification accuracy than some well-known feature selection methods.

Integrated Berth and Crane Scheduling Problem Considering Crane Coverage in Multi-Terminal Tidal Ports under Uncertainty

In this work, we study the integrated berth and crane scheduling problem in a tidal port with multiple terminals, considering the uncertainties, tides, maximum coverage of cranes and interference between cranes. For coping with the uncertainties, a certain number of randomly generated samples are used to evaluate the solutions, and slack variables are introduced to reduce the impact caused by the variation in vessel arrival and crane operational efficiency. A novel nonlinear mixed integer programming model is first formulated for the problem to minimize the sum of expectation and variance of costs under all samples. An improved adaptive genetic algorithm, combining a simulated annealing mechanism and greedy construction strategy, is developed and implemented by MATLAB. The feasibility and validity of the algorithm and the benefits of multi-terminal collaborative scheduling strategy under uncertainty are evaluated through numerical experiments. The results show that the algorithm can obtain feasible scheduling solutions with higher quality. Compared to the strategy that considers either the uncertainty or the multi-terminal collaborative mechanism, the resulting solution considering both can effectively reduce the cost and improve the competitiveness of the port.

Floating marine debris mitigation by vessel routing modeling and optimization considering carbon emission and travel time

Macro-marine debris becomes a global thorny issue to human beings and people are keeping trying every means to mitigate the risk of it. In this paper, we propose to use vessel and develop optimal vessel routing network to collect macro-marine debris on the nearshore surface. Due to the integrated force of winds and ocean currents, debris constantly changes floating location in the ocean. Advanced remote sensing technology and General National Oceanic and Atmospheric Administration (NOAA) Operational Modeling Environment (GNOME) software are employed to identify debris locations and track the drifting trajectory, respectively. In order to balance collecting efficiency and vessel emission reduction, we propose a bi-objective mixed integer nonlinear programming model for vessel routing to minimize travel time and carbon emission, considering time window at debris location, vessel capacity, low/medium/high vessel speed, and cost including carbon tax. A novel pheromone heuristic adaptive large neighborhood search (PHALNS ) algorithm combined with archived multi-objective simulated annealing (AMOSA) mechanism is developed to solve the proposed model. The Yangtze River Estuary region is taken as a numerical example to verify the proposed model and algorithm. Six criterions are introduced to evaluate the best collection time, and the proposed algorithm is also compared with NSGA II algorithm. The results show that low carbon emission and cost effective could be achieved simultaneously, and vessel speed has larger impact than the route scheme.

Improved Salp Swarm Algorithm with Simulated Annealing for Solving Engineering Optimization Problems

Swarm-based algorithm can successfully avoid the local optimal constraints, thus achieving a smooth balance between exploration and exploitation. Salp swarm algorithm (SSA), as a swarm-based algorithm on account of the predation behavior of the salp, can solve complex daily life optimization problems in nature. SSA also has the problems of local stagnation and slow convergence rate. This paper introduces an improved salp swarm algorithm, which improve the SSA by using the chaotic sequence initialization strategy and symmetric adaptive population division. Moreover, a simulated annealing mechanism based on symmetric perturbation is introduced to enhance the local jumping ability of the algorithm. The improved algorithm is referred to SASSA. The CEC standard benchmark functions are used to evaluate the efficiency of the SASSA and the results demonstrate that the SASSA has better global search capability. SASSA is also applied to solve engineering optimization problems. The experimental results demonstrate that the exploratory and exploitative proclivities of the proposed algorithm and its convergence patterns are vividly improved.

Time-Dependent Multi-Depot Heterogeneous Vehicle Routing Problem Considering Temporal–Spatial Distance

Aiming at the multi-depot heterogeneous vehicle routing problem under the time-dependent road network and soft time window, considering vehicle fixed cost, time window penalty cost and vehicle transportation cost, an optimization model of time-dependent multi-depot heterogeneous vehicle routing problem is established with the objective of minimizing distribution cost. According to the characteristics of the problem, a hybrid genetic algorithm with variable neighborhood search considering the temporal–spatial distance is designed. Customers are clustered according to the temporal–spatial distance to generate initial solutions, which improves the quality of the algorithm. The depth search capability of the variable neighborhood search algorithm is applied to the local search strategy of the genetic algorithm to enhance the local search capability of the algorithm. An adaptive neighborhood search number strategy and a new acceptance mechanism of simulated annealing are proposed to balance the breadth and depth required for population evolution. The validity of the model and algorithm is verified by several sets of examples of different scales. The research results not only deepen and expand the relevant research on vehicle routing problem, but also provide theoretical basis for logistics enterprises to optimize distribution scheme.

Optimization of SMOTE for imbalanced data based on AdaRBFNN and hybrid metaheuristics

Oversampling ratio N and the minority class’ nearest neighboring number k are key hyperparameters of synthetic minority oversampling technique (SMOTE) to reconstruct the class distribution of dataset. No optimal default value exists there. Therefore, it is of necessity to discuss the influence of the output dataset on the classification performance when SMOTE adopts various hyperparameter combinations. In this paper, we propose a hyperparameter optimization algorithm for imbalanced data. By iterating to find reasonable N and k for SMOTE, so as to build a balanced and high-quality dataset. As a result, a model with outstanding performance and strong generalization ability is trained, thus effectively solving imbalanced classification. The proposed algorithm is based on the hybridization of simulated annealing mechanism (SA) and particle swarm optimization algorithm (PSO). In the optimization, Cohen’s Kappa is used to construct the fitness function, and AdaRBFNN, a new classifier, is integrated by multiple trained RBF neural networks based on AdaBoost algorithm. Kappa of each generation is calculated according to the classification results, so as to evaluate the quality of candidate solution. Experiments are conducted on seven groups of KEEL datasets. Results show that the proposed algorithm delivers excellent performance and can significantly improve the classification accuracy of the minority class.

Time-dependent multi-depot green vehicle routing problem with time windows considering temporal-spatial distance

Reducing distribution costs is one of the effective ways for logistics enterprises to improve their core competitiveness. Aiming at the multi-depot vehicle routing problem under the time-varying road network, this paper proposes an integer programming model with the minimum total costs by comprehensively considering the fixed costs of vehicles, penalty costs on earliness and tardiness, fuel costs and the effects of vehicle speed, load and road gradient on fuel consumption. A hybrid genetic algorithm with variable neighborhood search is developed to solve the problem. In the algorithm, the temporal-spatial distance is introduced to cluster the customers to generate an initial population to improve the quality of the initial solution. Adaptive neighborhood search times strategy and simulated annealing inferior solution acceptance mechanism are used to balance the diversification and exploitation in the algorithm iteration process. Numerical results show that the model and algorithm we proposed are rather effective. The research results not only deepen and expand the vehicle routing problem (VRP) theory research, but also provide a scientific and reasonable method for logistics enterprises to make the vehicle scheduling plan.