人机协同创新脑风暴优化算法：解决两层车辆路线问题

182 浏览量更新于2024-08-29 收藏 161KB PDF 举报

本文主要探讨了一种基于人机协同创新思维优化算法的两阶段车辆路线问题解决方案。该研究论文名为"Human-computer Cooperative Brain Storm Optimization Algorithm for the Two-echelon Vehicle Routing Problem"，由薛明星、翟飞鸿、韩 Huang（通信作者）和李刚等学者共同完成，来自中国南方科技大学计算机科学与工程学院、佛山大学数学与大数据学院以及华南理工大学软件工程学院。论文的核心创新点在于提出了一种结合了计算机游戏中的智能和人类创造力的人机协同脑力激荡优化算法。在算法设计中，初始种群不仅包括通过计算机游戏得到的一些优质想法，而且在进化过程中，收敛操作和发散操作也利用不同玩家的解决方案来生成新的思想。这种设计使得算法能够充分利用人类的策略制定能力和计算机的高效计算能力，从而提升问题求解的效率和质量。具体而言，算法的工作流程包括以下几个关键步骤： 1. 利用计算机游戏中的数据和反馈，提供初始种群：游戏中的决策和行为模式可以作为算法的启发式信息，为优化过程提供良好的起点。 2. 人机协同的创新生成：在遗传/进化过程中，人类参与者可以根据自身的专业知识和经验对算法提出的候选解进行评估和改进，同时算法也会将优化结果反馈给人类，形成交互式的优化过程。 3. 整合策略发展能力：通过人类对算法的干预和指导，算法能够发展出更符合实际需求的解决方案策略，避免陷入局部最优。 4. 平衡收敛与发散：算法设计了动态调整的收敛和发散策略，确保在探索新解和保持现有优秀解之间找到最佳平衡，提高整体的寻优效果。 5. 应用到两阶段车辆路线问题：论文着重讨论了这种算法如何应用于具有挑战性的两阶段物流配送问题，旨在降低运输成本、提高效率并优化资源配置。这项研究为解决复杂的物流问题提供了一种新颖而有效的途径，展示了人机协同优化在实际问题中的潜力，为未来优化算法的设计和应用提供了有价值的参考。通过这种方式，我们可以期待在信息技术领域看到更多结合人工智能和人类智慧的创新解决方案。

Human-computer Cooperative Brain Storm

Optimization Algorithm for the Two-echelon

Vehicle Routing Problem

Xueming Yan

∗

, Zhifeng Hao

†

, Han Huang

‡

(Corresponding Author)* and Gang Li

∗

School of Computer Science and Engineering, South China University of Technology Guangzhou, China

†

School of Mathematics and Big Data, Foshan University, Foshan, China

‡

School of Software Engineering, South China University of Technology, Guangzhou, China, 510006,

Email:hhan@scut.edu.cn

School of Computer and Engineering, South China University of Technology, Guangzhou, China

Abstract—This paper presents a human-computer coop-

erative brain storm optimization algorithm, which is based

on an improved brain storm optimization algorithm with

human intelligence in computer game. In our algorithm,

the initial population is provided with some better ideas

obtained by computer game. Moreover, converging oper-

ation and diverging operation also employ the solutions

from different players to generate ideas during evolution

process. With the help of human-machine cooperation, our

algorithm, integrating strategy development capabilities of

players with brain storm optimization algorithm, is applied

to solve some complex optimized problems. We apply the

proposed method to two-echelon vehicle routing problem

to verify its effectiveness and usefulness.

I. INTRODUCTION

Lenat and Feigenbaum [1] ﬁrst proposed the idea

on human-machine cooperation in 1991. Recently, the

rapid growth of information systems has led to the wide

development of research on human-machine relationship.

Some new human-machine cooperation approaches are

necessary to be introduced by this trend [2]. According

to this idea, we attempt to take into account the inte-

gration of human strategy developing capabilities with

optimization algorithms for solving various scientiﬁc and

engineering problems. People exert great amounts of

problem-solving effort on playing computer games. For

example, protein structures prediction tasks have been

successfully through games [3]. However, it is not clear

whether more complex scientiﬁc problems can be solved

with the combination of human-direct computing and

traditional computational algorithms through multiplayer

games.

Many approaches are proposed to solve two-echelon

vehicle routing problems [4]. However, researchers still

believe that it is necessary to develop new methods

and technologies for optimizing, using the resources

presently available for reducing the impact of different

sources of nuisance [5]. Two-echelon vehicle routing

problems deal with how to ﬁnd a set of primary and

secondary routes, with which the demand of all cus-

tomers is satisﬁed, while the total distribution cost is

minimized [4]. Brain storm optimization (BSO) is a

new kind of swarm intelligence algorithm inspired by

human brainstorming in problem solving process [7].

The effectiveness and usefulness of BSO algorithms in

optimization problems can be illustrated in [8].

The framework of BSO algorithm is proposed by Shi.

There are ﬁve main operations of BSO [7]. The ﬁrst step

is to initialize a population of solutions uniformly and

randomly in searching space. Each individual is evaluat-

ed in the second step, and an evaluation value is got for

each solution as its ﬁtness. The third step is converging

operation which clusters all the individuals generated in

the current generation. Next step is diverging operator

that creates new individual by adding random noise to

the individual of one selected group or the individuals

of two groups, or the individual from even more groups.

Finally, the updating step updates individuals by creating

and selecting individuals in the population [9]. Recently

some approaches based on converging operation are

proposed in some literatures. Simple clustering algo-

rithms are employed in [11][12], which can accelerate

the speed of algorithm. Besides, self-determining number

of clusters approach in [13] is proposed to seek for higher

efﬁciency. Another convergent operation is proposed to

solve multi-objective optimization problems in [14][15].

Moreover, diverging operation is replaced based on the

basic BSO with closed-loop strategy in [16]. In this

2676

978-1-5090-0623-6/16/$31.00

2016 IEEE

下载后可阅读完整内容，剩余5页未读，立即下载

weixin_38564718

粉丝: 5
资源: 916

人机协同创新脑风暴优化算法：解决两层车辆路线问题

头脑风暴优化算法(Brain Storm Optimization Algorithm,BSO) 文档及Matlab算法

Distributed Consensus in Multi-vehicle Cooperative Control.rar

A Center Multi-swarm Cooperative Particle Swarm Optimization with Ratio and Prop

Efficient CPU-GPU cooperative computing for solving the subset-sum problem

The anti-windup control for hypersonic vehicles with input saturation based on improved cooperative optimization algorithm

A data management model for computer-supported cooperative work

Cooperative artificial bee colony algorithm for multi-objective RFID network planning

论文研究-Large-scale global optimization using cooperative coevolution with self-adaptive differential grouping.pdf

Multi-Agent Cooperative Bidding Games for Multi-Objective

AAM Two-on-one Cooperative Interception with Controllable Impact Time Difference

最新资源