基于排队模型的多技能呼叫中心人力需求问题研究

发布时间：2018-06-24 10:05

本文选题：呼叫中心 + 人力需求　；参考：《燕山大学》2016年博士论文

【摘要】：随着信息化大数据时代的到来,呼叫中心已经与人们的日常生活息息相关。政府,社会服务部门和企业,通过呼叫中心与顾客之间架起了一座沟通的桥梁。随着科技的进步,呼叫中心产业迅速发展,多技能成为了现代呼叫中心的主要特点。任何呼叫中心都希望在达到一定的顾客满意度的前提下把运营成本控制到最低,运营成本中最大一部分即人力成本,而计算出最优的人力需求量可以大大降低人力成本,因此,人力需求计算问题成为呼叫中心最关心的问题之一。人力需求计算的研究内容为在已知话务量的前提下,计算要达到一定服务水平需要的不同技能坐席人员的数量。本文主要针对N型和M型多技能呼叫中心的人力需求计算问题进行研究,主要的研究工作有以下四个方面:首先,分析了N型和M型多技能呼叫中心的排队系统模型。提出了新的状态空间划分方法,利用每个坐席组所处的不同状态,将系统的无穷多个状态划分成几个有限的状态集合,给出了状态集转移图,求出了状态集转移率,建立了系统的平衡方程并求出了状态集的稳态概率,基于以上结果进一步求出了系统服务水平的计算公式,并给出了数值算例进行分析。其次,研究了N型和M型多技能呼叫中心的人力需求计算问题。利用服务水平的计算公式,建立了求解最优坐席数目的人力需求计算模型,此模型是一个非线性整数规划问题。结合模型本身的特点,采用隐枚举法,利用Matlab软件编程,对模型进行了求解。通过数值算例验证了方法的可行性,分析了系统中参数的不同取值对结果的影响,并且进一步进行了实例分析。再次,研究了带有不耐烦顾客的N型和M型多技能呼叫中心人力需求问题。将顾客的不耐烦特性考虑到模型中是十分必要的,利用排队模型法求出了系统的性能指标和服务水平计算公式。进一步建立了人力需求计算模型,并将模型推广到了多种电话类型和多个坐席组的一般情况,采用了一种启发式算法,即蚁群算法对模型进行求解,给出了算法过程,并用Matlab软件编程,通过数值算例进一步分析了不耐烦因素对整个系统的影响,并进行了实例分析。最后,分析了带有排队信息提示的N型和M型多技能呼叫中心,即当顾客到达时通知顾客需要排队等待的时间。模型中有两种类型的电话到达,首先给出了顾客耐心无限的情况下两种类型电话等待时间的计算公式。进一步研究了带有不耐烦顾客的情况。基于马尔科夫链的方法,给出了等待时间的计算方法,并通过数值算例分析了增加排队信息提示对系统人力需求的影响。总之,本文对当前广泛应用的多技能呼叫中心的人力需求计算问题进行了研究。主要的创新性成果有:给出了N型和M型多技能呼叫中心服务水平的解析计算公式,建立了可以求出每个坐席组最优坐席数的人力需求计算模型并进行了求解和分析。进一步建立了带有不耐烦顾客的N型和M型多技能呼叫中心人力需求计算模型并进行了求解。此外,提出了带有排队信息提示的N型和M型多技能呼叫中心模型,并给出了等待时间的计算公式。本文的研究丰富了多技能呼叫中心的理论成果,可以为实际呼叫中心的人力需求计算问题提供一定的参考。
[Abstract]:With the advent of the era of large information data, the call center has been closely related to people's daily life. The government, social service departments and enterprises have built a bridge of communication between the call center and the customers. With the progress of science and technology, the call center industry has developed rapidly, and the multi skill has become the main feature of the modern call center. Every call center wants to control the operating cost to the lowest level on the premise of achieving a certain customer satisfaction. The largest part of the operation cost is the human cost, and the calculation of the optimal human demand can greatly reduce the human cost. Therefore, the problem of human demand calculation becomes one of the most concerned problems in the call center. The research content of calculation is to calculate the number of different skilled seats required by a certain service level under the premise of known traffic volume. This paper mainly focuses on the human demand calculation of N and M multi skill call centers. The main research work has four aspects: first, it analyzes the N type and the M type. A new method of state space division is proposed. Using the different state of each seat group, the infinitely many states of the system are divided into several finite state sets, the state set transfer graph is given, the transfer rate of the state set is obtained, the equilibrium equation of the system is established and the state set is obtained. Based on the above results, the calculation formula of the system service level is further derived based on the above results, and the numerical example is given. Secondly, the human demand calculation problem of N and M multi skill call centers is studied. The model is a nonlinear integer programming problem. Combining the characteristics of the model itself, the implicit enumeration method is used and the Matlab software is used to solve the model. The feasibility of the method is verified by a numerical example. The influence of the different values of the parameters on the results is analyzed, and the example analysis is further carried out. Thirdly, the study is made. The human demand problem of N and M multi skill call centers with impatient customers. It is necessary to consider the customer's impatience characteristics in the model. The queuing model method is used to find the system performance index and the service level calculation formula. The human requirement calculation model is further established and the model is extended to a variety of telephone. A heuristic algorithm, ant colony algorithm, is used to solve the model, and the algorithm process is given. The effect of impatience factors on the whole system is further analyzed by a numerical example by Matlab software. Finally, an analysis is made to analyze the queuing information. The N and M multi skill call centers, when the customer arrives, notify the customer the time to wait. There are two types of telephone arrival in the model. First, a formula for calculating the waiting time of two types of phone calls is given in the case of the patient's impatience. Method, the calculation method of waiting time is given, and a numerical example is used to analyze the effect of increasing the queuing information on the human needs of the system. In a word, this paper studies the human demand calculation problem of the widely used multi skill call center. The main innovative results are as follows: the N and M multi skill call centers are given. The analytical calculation formula of the service level is established, and the human demand calculation model of the optimal seat number for each seat group is established and the solution and analysis are carried out. The human demand calculation model of N and M multi skill call centers with impatient customers is further established and the solution is solved. In addition, the queuing information prompts are proposed. The model of N and M multi skill call center and the calculation formula of waiting time are given. The research of this paper enriches the theoretical results of the multi skill call center, and can provide some reference for the calculation of human demand in the actual call center.
【学位授予单位】：燕山大学
【学位级别】：博士
【学位授予年份】：2016
【分类号】：TN99

【相似文献】