呼叫中心排班人员数目决策的实证研究
发布时间:2018-09-08 18:18
【摘要】: 随着移动电话业务的急剧增长,电信公司为了保持或提升其竞争力,建立电话呼叫中心为顾客提供24小时的热线服务。本文研究的某电信呼叫中心以提供基于电话服务的内呼业务为主,负责对公司推出的各类手机卡产品及电信服务提供话费咨询、业务受理、投诉处理等服务项目,并协调公司其它部门及时协调、解决顾客问题。作为以提供内呼业务为主的呼叫中心而言,迫于竞争对手的压力以及买方话务到达量的模式,导致在电话队列中等待时间过长或者根本不能顺利接入服务系统的顾客,很可能会转向企业的竞争对手并成为其客户。因此呼叫中心的拨通率,即保证顾客随时可以方便地接入呼叫中心接受服务,成为呼叫中心运营的基本目标。 作为一个追求质量与效率平衡的随机服务系统,呼叫中心拨通率指标的实现有赖于在有限的成本基础上,根据每天不同时段变化的话务到达量来相应地安排客服代表数目。国内电信呼叫中心在进行人员数目求解时,通常采用ErlangC排队公式,但是该排队模型的成立需要基于许多假设条件之上。而现代呼叫中心的运营特点、话务到达量变化以及顾客放弃行为的不可忽略,都使得原有模型远远不能满足实际人员排班需要。本文通过对某电信呼叫中心实际运营数据的分析,对顾客放弃行为的影响进行探讨,并对ErlangC和ErlangA两种排队模型在预测人员需求数目方面的差异进行对比和分析,验证了运用ErlangA模型进行人员数目计算的优越性。并对排队模型的三大输入参数:话务到达率、客服代表的服务时间以及顾客排队等候的耐心程度进行深入分析。通过Erlang排队模型最终计算出来的是当量人员数目。为了弥补话务高峰期服务系统的随机变异性,本文又提出用改进的平方根安全人员原理进行安全人员数目的补充,以确保在高负荷利用率的情况下仍然能够保证呼叫中心拨通率的实现。 本研究的创新之处在于,基于呼叫中心实际运营数据的分析而提出呼叫中心人员需求数目的新求解模型;证明国内某呼叫中心的客服代表服务时长服从对数正态分布,并以此进行均值和置信区间的计算;最后运用ErlangA模型解决质量与效率平衡系统下的安全人员数目求解问题。
[Abstract]:With the rapid growth of mobile phone business, in order to maintain or enhance their competitiveness, telecom companies set up call centers to provide customers with 24-hour hotline service. The telecom call center studied in this paper mainly provides internal call service based on telephone service. It is responsible for providing telephone fee consultation, business acceptance, complaint handling and other service items to all kinds of mobile phone card products and telecommunication services launched by the company. Coordinate with other departments to solve customer problems. As a call center that mainly provides internal call services, customers who wait too long in a telephone queue or simply cannot access the service system are subjected to the pressure of their competitors and the mode of arrival of the buyer's traffic. It is likely to turn to its competitors and become its customers. Therefore, the dialing rate of the call center, that is, to ensure that the customer can easily access the call center to receive services at any time, has become the basic target of the call center operation. As a stochastic service system which pursues the balance of quality and efficiency, the implementation of call center dialing rate index depends on arranging the number of customer service representatives on the basis of the limited cost and according to the traffic arrival quantity varying at different times of the day. The ErlangC queuing formula is usually used in the calculation of the number of people in the domestic telecom call center, but the establishment of the queuing model needs to be based on many assumptions. However the operation characteristics of modern call center the change of traffic arrival volume and the abandon behavior of customers all make the original model far from meeting the needs of actual personnel scheduling. Based on the analysis of the actual operation data of a telecom call center, this paper probes into the influence of customer abandonment behavior, and compares and analyzes the difference between ErlangC and ErlangA queuing models in predicting the number of personnel demand. The superiority of using ErlangA model to calculate the number of personnel is verified. The three input parameters of the queuing model, namely, the traffic arrival rate, the service time of the customer service representative and the patient degree of the customer queuing are analyzed. The number of equivalent personnel is calculated by Erlang queuing model. In order to make up for the random variability of traffic peak service system, this paper proposes to use the improved square root security personnel principle to supplement the number of security personnel. To ensure that the call center dialing rate can still be achieved under high load utilization. The innovation of this study is that, based on the analysis of the actual operation data of the call center, a new solution model for the number of call center personnel is proposed, and it is proved that the long service of customer service representative service in a domestic call center is logarithmic normal distribution. Finally, the ErlangA model is used to solve the problem of the number of security personnel in the system of quality and efficiency balance.
【学位授予单位】:天津大学
【学位级别】:硕士
【学位授予年份】:2006
【分类号】:F626;C934
本文编号:2231311
[Abstract]:With the rapid growth of mobile phone business, in order to maintain or enhance their competitiveness, telecom companies set up call centers to provide customers with 24-hour hotline service. The telecom call center studied in this paper mainly provides internal call service based on telephone service. It is responsible for providing telephone fee consultation, business acceptance, complaint handling and other service items to all kinds of mobile phone card products and telecommunication services launched by the company. Coordinate with other departments to solve customer problems. As a call center that mainly provides internal call services, customers who wait too long in a telephone queue or simply cannot access the service system are subjected to the pressure of their competitors and the mode of arrival of the buyer's traffic. It is likely to turn to its competitors and become its customers. Therefore, the dialing rate of the call center, that is, to ensure that the customer can easily access the call center to receive services at any time, has become the basic target of the call center operation. As a stochastic service system which pursues the balance of quality and efficiency, the implementation of call center dialing rate index depends on arranging the number of customer service representatives on the basis of the limited cost and according to the traffic arrival quantity varying at different times of the day. The ErlangC queuing formula is usually used in the calculation of the number of people in the domestic telecom call center, but the establishment of the queuing model needs to be based on many assumptions. However the operation characteristics of modern call center the change of traffic arrival volume and the abandon behavior of customers all make the original model far from meeting the needs of actual personnel scheduling. Based on the analysis of the actual operation data of a telecom call center, this paper probes into the influence of customer abandonment behavior, and compares and analyzes the difference between ErlangC and ErlangA queuing models in predicting the number of personnel demand. The superiority of using ErlangA model to calculate the number of personnel is verified. The three input parameters of the queuing model, namely, the traffic arrival rate, the service time of the customer service representative and the patient degree of the customer queuing are analyzed. The number of equivalent personnel is calculated by Erlang queuing model. In order to make up for the random variability of traffic peak service system, this paper proposes to use the improved square root security personnel principle to supplement the number of security personnel. To ensure that the call center dialing rate can still be achieved under high load utilization. The innovation of this study is that, based on the analysis of the actual operation data of the call center, a new solution model for the number of call center personnel is proposed, and it is proved that the long service of customer service representative service in a domestic call center is logarithmic normal distribution. Finally, the ErlangA model is used to solve the problem of the number of security personnel in the system of quality and efficiency balance.
【学位授予单位】:天津大学
【学位级别】:硕士
【学位授予年份】:2006
【分类号】:F626;C934
【引证文献】
相关期刊论文 前1条
1 苏晓东;梁艳;;多业务呼叫中心系统坐席分配问题研究[J];信息技术;2010年08期
相关硕士学位论文 前1条
1 黄小斌;基于CTI技术的社区呼叫中心的研究与实现[D];南昌大学;2009年
,本文编号:2231311
本文链接:https://www.wllwen.com/guanlilunwen/tongjijuecelunwen/2231311.html