联系我们

论文查询结果	返回搜索

论文编号：	13216
作者编号：	2120202797
上传时间：	2022/6/7 16:26:42
中文题目：	混合式多技能呼叫中心坐席排班优化研究
英文题目：	Research on Shift Scheduling optimization of hybrid multi-skill call center
指导老师：	王谦老师
中文关键字：	呼叫中心；混合渠道；多技能；排班优化； 启发式算法
英文关键字：	Call Center; Mixed channels; Multi skill; Scheduling optimization; Heuristic algorithm
中文摘要：	呼叫中心作为服务供应商与企业客户关联的重要纽带，长期是现代呼叫关系管理的前沿领域。在以往的多数研究中，呼叫中心被简单定义为通过电话语音交互方式，支持企业客户关系管理的服务组织。但在现代信息技术的加持和推动下，呼叫中心的服务模式已发生深刻变革，传统语音型呼叫中心正向多渠道协同型呼叫中心转型。当前基于即时通讯技术的文字交互逐渐成为企业青睐的通信方式，全球超过40%的呼叫中心在运营实践中使用即时聊天交互，现代呼叫中心正逐步进入传统语音处理和基于文字通讯的文字处理的混合服务时代，研究多渠道多技能的现代呼叫中心具备现实意义和实用价值。 本文针对混合语音与文字渠道的多技能坐席排班问题进行研究，基于季节性差分自回归移动平均模型（SARIMA）预测语音、文字渠道呼入波动，在构建多技能呼叫中心仿真模型的基础上，提出两类改进全局优化性能的粒子群算法，并结合OCBA算法对坐席排班方案进行优化。本文在所提出的仿真优化方法基础上，深入探究影响混合式呼叫中心班次人力配比的关键因素，在分析问题优化结构规律的基础上提出适用的启发式算法HSHA，并以某呼叫中心的实例数据对模型及算法的有效性进行验证，以期有效降低人力排班求解过程的时间成本。 实证研究表明，本文所提出的两类改进的粒子群算法，调整了粒子迭代进化所采用的速度更新公式，一定程度上避免了传统粒子群算法容易陷入局部寻优的问题，提高了算法的全局优化能力，支持获得企业实际场景的优化排班方案。本文比对两类改进算法的排班优化性能，实验结果表明，当精英子集包含的方案个数m在合适区间内，改进的IPSO2-m算法较IPSO1算法表现出更好的优化效果和算法稳定性。本文研究的混合式呼叫中心排班启发式算法HSHA，在优化效果不劣于仿真优化方法的基础上，极大提升排班优化方案的获取效率，具备较高的理论价值和实用性。
英文摘要：	As an important link between service providers and enterprise customers, call center has long been the forefront of modern call relationship management. In most previous studies, call center is simply defined as a service organization that supports enterprise customer relationship management through telephone voice interaction. However, with the support and promotion of modern information technology, the service mode of call center has undergone profound changes. The traditional voice call center is transforming into a multi-channel collaborative call center. At present, text interaction based on instant messaging technology has gradually become a preferred communication mode for enterprises. More than 40% of the world's call centers use instant chat interaction in operation practice. Modern call centers are gradually entering the era of hybrid service of traditional voice processing and text processing based on text communication, The study of multi-channel and multi skill modern call center has practical significance and practical value. This thesis studies the multi skill seat scheduling problem of mixed voice and text channels. Based on the Seasonal Autoregressive Integrated Moving Average model (SARIMA), the incoming call fluctuation of voice and text channels is predicted. Based on constructing the multi skill call center simulation model, two kinds of improved particle swarm optimization algorithms to improve the global optimization performance are proposed, and the seat scheduling scheme is optimized combined with OCBA algorithm. Based on the proposed simulation optimization method, this thesis deeply explores the key factors affecting the shift manpower ratio of hybrid call center, puts forward the applicable heuristic algorithm HSHA based on the analysis of the structural law of problem optimization, and verifies the effectiveness of the model and algorithm with the empirical process of a call center, in order to effectively reduce the time cost of manpower scheduling solution process. Empirical research shows that the two improved particle swarm optimization algorithms proposed in this thesis adjust velocity updating formula used in particle iterative evolution, avoid the problem that the traditional particle swarm optimization algorithm is easy to fall into local optimization to a certain extent, improve the global optimization ability of the algorithm, and support the optimal scheduling scheme of the actual scene of the enterprise. This thesis compares the scheduling optimization performance of two kinds of improved algorithms. The experimental results show that when the number of schemes m contained in the refined subset is within the appropriate interval, the improved algorithm IPSO2-m shows better optimization effect and algorithm stability than IPSO1. The hybrid call center scheduling heuristic algorithm HSHA studied in this thesis greatly improves the acquisition efficiency of scheduling optimization scheme on the basis that the optimization effect is not inferior to the simulation optimization method, and has high theoretical value and practicability.
查看全文：	预览  下载(下载需要进行登录)