计算机科学

计算机科学 ›› 2025, Vol. 52 ›› Issue (10): 33-49.doi: 10.11896/jsjkx.250500086

• 数智赋能金融科技前沿 • 上一篇    下一篇

融合机器学习预测和水波优化算法求解银行在线客服调度问题

卢雪琴1, 谢歙铖2, 唐燕3, 陈世昆3, 刘仰光3   

  1. 1 浙江工业大学计算机科学与技术学院 杭州 310023
    2 杭州师范大学信息科学与技术学院 杭州 311121
    3 宁波财经学院金融与信息学院 浙江 宁波 315175
  • 收稿日期:2025-05-21 修回日期:2025-07-20 出版日期:2025-10-15 发布日期:2025-10-14
  • 通讯作者: 刘仰光(liuyangguang@nbufe.edu.cn)
  • 作者简介:(luxueqin@nbufe.edu.cn)
  • 基金资助:
    浙江省基础公益研究计划项目(LGF22G020002);浙江省教育厅高等学校访问学者专业发展项目(FX2023059);宁波市自然科学基金(2023J060)

Integration of Machine Learning Prediction and Water Wave Optimization for Online Customer Service Representatives Scheduling in Bank Contact Centers

LU Xueqin1, XIE Xicheng2 , TANG Yan3, CHEN Shikun3, LIU Yangguang3   

  1. 1 College of Computer Science and Technology,Zhejiang University of Technology,Hangzhou 310023,China
    2 School of Information Science and Technology,Hangzhou Normal University,Hangzhou 311121,China
    3 College of Finance & Information,Ningbo University of Finance & Economics,Ningbo,Zhejiang 315175,China
  • Received:2025-05-21 Revised:2025-07-20 Online:2025-10-15 Published:2025-10-14
  • About author:LU Xueqin,born in 1980,Ph.D candidate,associate professor,is a member of CCF(No.88689G).Her main research interest includes intelligent optimization algorithms and applications.
    LIU Yangguang,born in 1975,Ph.D,professor,master's supervisor,Zhejiang Province 151 Talent Program awardee,is a member of CCF(No.20163M).His main research interests include multi-scale data intelligence,AI and applied optimization.
  • Supported by:
    Zhejiang Provincial Basic Public Welfare Research Program(LGF22G020002),Zhejiang Provincial Department of Education College Visiting Scholars Professional Development Program(FX2023059) and Ningbo Natural Science Foundation(2023J060).

PDF (PC)

1

摘要： 在线客服的调度是银行客服中心的核心运营环节。高效的客服调度通过合理的人员配置与排班,确保客户能够及时获得服务,从而提升客户体验。然而,客户请求到达的随机性以及客服技能水平差异,使得在线客服调度问题变得复杂。对此,综合考虑客服技能等级、客户类型多样性以及匹配需求等因素,构建了一个以客户等待时间和运营成本最小化为优化目标的混合整数线性规划模型。针对客户需求的不确定性可能会导致客户需求和客服匹配困难,以及该问题在高维解空间中的求解复杂性,提出了一种融合机器学习预测与水波优化算法的混合方法来求解该客服调度问题。在该方法中,采用长短期记忆神经网络对客户到达量进行预测,充分捕捉其时间序列依赖性及外部因素的影响。对于客服调度的混合整数规划模型,则通过一种结合强化学习Q-learning的水波优化算法进行高效求解。以浙江泰隆银行宁波分行在线客服中心的真实数据为基础进行实验,结果表明,所提方法在运营成本控制方面显著优于对比方法。进一步的灵敏度分析揭示:当预测准确率低于90%时,因客户到达量的不确定性,调度成本与客户等待时长均显著上升;而当预测准确率达到或超过90%后,系统性能的提升趋于平缓。这些发现不仅验证了高精度预测对调度效果的显著影响,还为实际应用中平衡预测模型复杂度与调度效率提供了理论基础和实践指导。

关键词: 在线客服调度, LSTM神经网络, 强化学习, 水波优化算法

Abstract: Online customer service representative scheduling is a crucial component of operational management in a bank's contact center.Optimized staffing and shift scheduling of customer service representatives ensure prompt customer responses,which significantly improves service efficiency and customer satisfaction.However,factors such as the uncertainty of customer request arrivals and variations in customer service skill levels make the online service representatives scheduling problem complex.Considering the practical challenges such as customer service representative skill levels,diversity of customer types,and matching requirements,this study proposes a mixed-integer linear programming model with the optimization objectives of minimizing custo-mer waiting time and operational costs,and also presents a hybrid machine learning and water wave optimization(WWO) method to effectively solve the online customer service scheduling problem.In this method,a forecasting model based on long short-term memory neural networks is employed to predict customer arrival volumes,and this model can capture both time series dependencies and the influence of external factors.For the mixed-integer programming model of representative scheduling,WWO combining reinforcement learning Q-learning is used for efficient solution.This method leverages Q-learning to adaptively optimize neighborhood selection,enhancing the efficiency and quality of solutions.Based on real data from the contact center of Zhejiang Tailong Bank's Ningbo branch,the experimental results show that the proposed method significantly outperforms comparative methods in terms of operational cost control.Further,sensitivity analysis reveals that when forecast accuracy drops below 90%,customer arrival uncertainty markedly increases scheduling costs and customer waiting times.Conversely,when accuracy reaches or exceeds 90%,performance improvements stabilize.These findings highlight the critical role of high-precision forecasting in effective scheduling and provide theoretical insights and practical guidance for balancing forecasting model complexity with scheduling efficiency in real-world applications.

Key words: Online customer service representatives scheduling,LSTM neural network,Reinforcement learning,Water wave optimization

中图分类号: 

  • TP181
[1]DAI T,ZHAO X.A queuing model for 1-n interactive service mode in online customer service systems[J].Journal of Systems &Management,2021,30(3):562-570.
[2]YU M,ZHAO Y,CHANG C G,et al.Fluid models for customer service web chat systems with interactive automated service[J].Flexible Services and Manufacturing Journal,2023,35(2):572-598.
[3]MORTEZA S,OMAR K H,ELIZABETH C.Past,present and future of contact centers:a literature review[J].Business Process Management Journal,2017,23(3):574-597.
[4]DAI T,YU T T,ZHAO X.Decision strategy of single serveronline customer service with variable service rates[C]//2019 International Conference on Industrial Engineering and Systems Management(IESM).IEEE,2019:1-6.
[5]SU Q,ZHAO F.Research on staff scheduling algorithm formulti-skill call center[J].Systems Engineering-Theory & Practice,2015,35(1):239-246.
[6]ERNST A T,JIANG H,KRISHNAMOORTHY M,et al.Staff scheduling and rostering:A review of applications,methods and models[J].European Journal of Operational Research,2004,153(1):3-27.
[7]CHAN W,ANH T T,L'ECUYER P,et al.Two-stage chance-constrained staffing with agent recourse for multi-skill call centers[C]//2016 Winter Simulation Conference(WSC).IEEE,2016:3189-3200.
[8]ROBBINS T R,HARRISON T P.A stochastic programmingmodel for scheduling call centers with global Service Level Agreements[J].European Journal of Operational Research,2010,207(3):1608-1619.
[9]TÜRKER T,DEMIRIZ A A.An MIP model to schedule thecall center workforce and organize the breaks[C]//Innovations Through Mathematical And Statistical Research:Proceedings of the 2nd International Conference on Mathematical Sciences and Statistics(ICMSS2016).2016:020048.
[10]ÁLVAREZ E,FERRER J C,MUÑOZ J C,et al.Efficient shift scheduling with multiple breaks for full-time employees:A retail industry case[J].Computers & Industrial Engineering,2020,150:106884.
[11]VASILYEV I,USHAKOV A V,ARKHIPOV D,et al.Fastheuristics for a staff scheduling problem with time interval demand coverage[M]//Mathematical Optimization Theory and Operations Research.Cham:Springer,2024:212-228.
[12]WU X Q,ZHOU S H.Sequencing and scheduling appointments on multiple servers with stochastic service durations and customer arrivals[J].Omega,2022,106:102523.
[13]PANDEY P,GAJJAR H,SHAH B J.Determining optimalworkforce size and schedule at the retail store considering overstaffing and understaffing costs[J].Computers & Industrial Engineering,2021,161:107656.
[14]MA Y,LIU L L.Solving call center agent scheduling problemthrough improved adaptive genetic algorithm[C]//2012 Fifth International Symposium on Computational Intelligence and Design.2012:27-30.
[15]YAN L J,KONG X M,WANG N,et al.Research on artificial seat scheduling algorithm based on genetic algorithm[C]//Proceedings of the 2024 8th International Conference on Control Engineering and Artificial Intelligence.2024:26-29.
[16]REKIK M,CORDEAU J F,SOUMIS F.Implicit shift scheduling with multiple breaks and work stretch duration restrictions[J].Journal of Scheduling,2010,13(1):49-75.
[17]WANG H B,ALIDAEE B,ORTIZ J,et al.The multi-skilled multi-period workforce assignment problem[J].International Journal of Production Research,2021,59(18):5477-5494.
[18]NTHIMO L,MOKOENA T,MODUPE A,et al.Call centreshift schedule optimisation using local search heuristics[C]//2021 IEEE Africon.IEEE,2021:1-6.
[19]DHIBAR S,JAIN M.Metaheuristic optimization and strategicbehavior of Markovian vacation queue with retrial policy:application to virtual call center[J].Evolutionary Intelligence,2025,18(1):13.
[20]XU Y,WANG X L.An artificial bee colony algorithm for sche-duling call centres with weekend-off fairness[J].Applied Soft Computing,2021,109:107542.
[21]NIU G,VADIVELOO J,XU M N.Agent-based queuing model for call center forecasting and management optimization[M]//Advances in Business and Management Forecasting.Emerald Publishing Limited,2019:121-133.
[22]DE MOURA A F,DE ARAU＇JO C M,NAPOLITANO D M R.et al.Optimization of operational costs of Call centers employing classification techniques[J].Research,Society and Development,2020,9(11):e86691110491.
[23]HENRY C,VISHWA K,DAVID D,et al.Forecasting call center arrivals using temporal memory networks and gradient boosting algorithm[J].Expert Systems with Applications,2023,224:119983.
[24]NING C,YOU F Q.Optimization under uncertainty in the era of big data and deep learning:When machine learning meets mathe-matical programming[J].Computers & Chemical Engineering,2019,125:434-448.
[25]ZHENG Y J,XIE X C,ZHANG Z Y,et al.Deep reinforcement learning assisted memetic scheduling of drones for railway catenary deicing[J].Swarm and Evolutionary Computation,2024,91:101719.
[26]XU Y,LI X B,MENG X P.A Q-learning based iterated local search algorithm for human-UAV cooperation in restoring transmission network[J].Expert Systems with Applications,2024,252:124200.
[27]RAKSHIT P,KONAR A,BHOWMIK P,et al.Realization of anAdaptive Memetic Algorithm Using Differential Evolution and Q-Learning:A Case Study in Multirobot Path Planning[J].IEEE Transactions on Systems,Man,and Cybernetics:Systems,2013,43(4):814-831.
[28]JOUINI O,POT A,KOOLE G,et al.Online scheduling policies for multiclass call centers with impatient customers[J].Euro-pean Journal of Operational Research,2010,207(1):258-268.
[29]IBRAHIM R,YE H,L'ECUYER P,et al.Modeling and forecasting call center arrivals:A literature survey and a case study[J].International Journal of Forecasting,2016,32(3):865-874.
[30]SHEN H P,HUANG J H.Interday Forecasting and Intraday Updating of Call Center Arrivals[J].Manufacturing & Service Operations Management,2008,10(3):391-410.
[31]BARROW D,KOURENTZES N.The impact of special days in call arrivals forecasting:A neural network approach to modelling special days[J].European Journal of Operational Research,2018,264(3):967-977.
[32]MANNO A,ROSSI F,SMRIGLIO S,et al Comparing deep and shallow neural networks in forecasting call center arrivals[J].Soft Computing,2023,27(18):12943-12957.
[33]HU Y,CATO K D,CHAN C W,et al.Use of real-time information to predict future arrivals in the emergency department[J].Annals of Emergency Medicine,2023,81(6):728-737.
[34]HUBER J,STUCKENSCHMIDT H.Daily retail demand fore-casting using machine learning with emphasis on calendric special days[J].International Journal of Forecasting,2020,36(4):1420-1438.
[35]HUANG R J,WEI C J,WANG B H,et al.Well performanceprediction based on Long Short-Term Memory(LSTM) neural network[J].Journal of Petroleum Science and Engineering,2022,208:109686.
[36]ZHENG Y J,LU X Q,DU Y C,et al.Water wave optimization for combinatorial optimization:Design strategies and applications[J].Applied Soft Computing,2019,83:105611.
[37]SARDJONO W,PRIATNA W,NUGROHO D S,et al.Genetic algorithm implementation for application of shifting work scheduling system[M].ICIC International Association,2021.
[38]ZHANG L,WANG L,ZHENG D Z.An adaptive genetic algorithm with multiple operators for flowshop scheduling[J].The International Journal of Advanced Manufacturing Technology,2006,27(5/6):580-587.
[39]AKJIRATIKARL C,YENRADEE P,DRAKE P R.PSO-based algorithm for home care worker scheduling in the UK[J].Computers & Industrial Engineering,2007,53(4):559-583.
[40]AKJIRATIKARL C,YENRADEE P,DRAKE P R.An Im-proved Particle Swarm Optimization Algorithm for Care Worker Scheduling[J].Industrial Engineering & Management Systems,2008,7(2):171-181.
[41]ZANAZZO E,CESCHIA S,DOVIER A,et al.Solving the medical student scheduling problem using simulated annealing[J].Journal of Scheduling,2024,28:233-246.
[42]SHAHNAZARI P,TAVAKKOLI-MOGHADDAM R,AZARK-ISH M,et al.A differential evolution algorithm developed for a nurse scheduling problem[J].The South African Journal of Industrial Engineering,2012,23(3):68.
[43]PUNYAKUM V,NITISIRI K,SETHANAN K,et al.A self-adaptive differential evolution for the technician routing and scheduling problem[J].Engineering and Applied Science Research,2023,50(3):262-269.
[44]LU X Q,WU C X,YANG X H,et al.Adapted water wave optimization for integrated bank customer service representative scheduling[J].International Journal of Production Research,2023,61(1):320-335.
Viewed
Full text


Abstract

Cited
  Shared   
  Discussed   
No Suggested Reading articles found!
渝ICP备16013121号-4
地址：重庆市渝北区洪湖西路18号    邮编：401121  
电话：023-63500828（编辑部） 023-67039612（会议/专题） 023-67039623（财务/发票）
E-mail：jsjkx12@163.com
本系统由北京玛格泰克科技发展有限公司设计开发