Skill-based Routing Research Articles

A survey on skill-based routing with applications to service operations management

Service systems often feature multiple classes of customers with different service needs and multiple pools of servers with different skillsets. How to efficiently match customers of different classes with servers of different skillsets is of great importance to the management of these systems. In this survey, we review works on skill-based routing in queues. We first summarize key insights on routing/scheduling policies developed in the literature. We then discuss complications brought by modern service operations management problems, particularly healthcare systems. These complications stimulate a growing body of literature on new modeling and analysis tools. Lastly, we provide additional numerical experiments to highlight the complex nature of a routing problem motivated from hospital patient-flow management, and provide some useful intuition to develop good skill-based routing policies in practice. Our goal is to provide a brief overview of the skill-based routing research landscape and to help generate interesting research ideas.

COMPETENCIAS COMUNICATIVAS EN TÍTULOS UNIVERSITARIOS NO LINGÜÍSTICOS

Author/s
 Oksana PolyakovaRuzana Galstyan-SargsyanUniversidad Católica de Valenica, Spain
 
 ABSTRACT
 As labour markets continue to change rapidly, the learning environments continue to be unchallenged in the provision of new professionals across the world. This change has compelled the need for students to attain communicative competences in non-linguistic institutions of higher education. In order to address this issue, we propose a novel approach to communicative competences integration into the curriculum development of a specialised foreign language course. For this purpose, we combine students’ needs analysis with a personalised English for specific purposes (ESP) course design that motivates them to learn more about technical and professional environments through a skill-based routing. Additionally, we provide a scale for measuring training outcomes as well as suggest first results from the piloting phase of the teaching experience. The findings of the study highlight usefulness of the customised competence training experience.

Flexible Queueing Architectures

We study a multiserver model with n flexible servers and n queues, connected through a bipartite graph, where the level of flexibility is captured by an upper bound on the graph’s average degree, dn. Applications in content replication in data centers, skill-based routing in call centers, and flexible supply chains are among our main motivations. We focus on the scaling regime where the system size n tends to infinity, while the overall traffic intensity stays fixed. We show that a large capacity region and an asymptotically vanishing queueing delay are simultaneously achievable even under limited flexibility (dn ≪ n). Our main results demonstrate that, when dn ≫ ln n, a family of expander-graph-based flexibility architectures has a capacity region that is within a constant factor of the maximum possible, while simultaneously ensuring a diminishing queueing delay for all arrival rate vectors in the capacity region. Our analysis is centered around a new class of virtual-queue-based scheduling policies that rely on dynamically constructed job-to-server assignments on the connectivity graph. For comparison, we also analyze a natural family of modular architectures, which is simpler but has provably weaker performance. The online appendix is available at https://doi.org/10.1287/opre.2017.1620 .

A flexible architecture for call centers with skill-based routing

We focus on architectures with limited flexibility for multi-skill call centers. The context is that of call centers with asymmetric parameters: unbalanced workload, different service requirements, a predominant customer type, unbalanced abandonments and high costs of cross-training. The most well-known architectures with limited flexibility such as chaining fail against such asymmetry. We propose a new architecture referred to as single pooling with only two skills per agent and we demonstrate its efficiency. Using simulation, we conduct a comprehensive comparison between this architecture and chaining. As a function of the various system parameters, we delimit the regions where either chaining or single pooling is the best. Single pooling leads to a better performance than chaining while being less costly under various situations of asymmetry: asymmetry in the number of arrivals, in the service durations, in the variability of service times, or in the service level requirements. It is also shown that these observations are more apparent for situations with a large number of skills, or for those with a large call center size.

Queueing system topologies with limited flexibility

We study a multi-server model with n flexible servers and rn queues, connected through a fixed bipartite graph, where the level of flexibility is captured by the average degree, d(n), of the queues. Applications in content replication in data centers, skill-based routing in call centers, and flexible supply chains are among our main motivations. We focus on the scaling regime where the system size n tends to infinity, while the overall traffic intensity stays fixed. We show that a large capacity region (robustness) and diminishing queueing delay (performance) are jointly achievable even under very limited flexibility (d(n) l n). In particular, when d(n) gg ln n , a family of random-graph-based interconnection topologies is (with high probability) capable of stabilizing all admissible arrival rate vectors (under a bounded support assumption), while simultaneously ensuring a diminishing queueing delay, of order ln n/ d(n), as n-> ∞. Our analysis is centered around a new class of virtual-queue-based scheduling policies that rely on dynamically constructed partial matchings on the connectivity graph.

Models and valid inequalities to asymmetric skill-based routing problems

Given a directed network, where a set of skills S j is required to serve each node j , and given a set of available technicians, each one having abilities described through a set of skills, we study the problem of defining the tour of each technician in such a way that each service requirement is fulfilled by exactly one technician, and proper skill constraints are satisfied. These constraints state that the service requirement at node j can be operated by any technician having, at least, all the skills in S j . Given travelling costs for the arcs of the network, which are technician dependent, we want to determine a set of tours of minimum cost which satisfy the skill constraints. This problem, named asymmetric skill VRP (A-SKVRP), originates from a real application arising in Field Service. Also, it is strongly related to Home Care applications. Starting from the preliminary results in Cappanera et al. (Network optimization. 5th international conference, INOC 2011, Hamburg, June 2011. Lecture notes in computer science, Springer, Berlin, vol 6701, pp 354–364, 2011 ), where a special case is addressed, we propose a hierarchy of three ILP flow-based models for A-SKVRP, which is based on increasing levels of disaggregation of the flow variables used to prevent subtours. The computational results of a first pool of experiments show that the LP bounds produced by the most disaggregated model are very tight, but at the expense of very large computational times in most cases. This becomes critical as the number of involved nodes and skills increases. Therefore, in the second part of our study, we enhance the LP bounds obtained with the weaker models by adding, in a cutting plane fashion, cut constraints. As a result of this enhancement, the intermediate model proves to be competitive against the most disaggregated model in computing tight lower bounds, especially as the number of the nodes and of the skills increases. This model combined with the proposed valid inequalities thus qualifies as an efficient tool to address skill-based routing problems arising in application contexts such as Field Service and Home Care.

Contributors

Contributors

Benefits of cross-training in a skill-based routing contact center with priority queues and impatient customers

Customer contact centers that provide different types of services to customers who place phone calls or send e-mail messages are studied. Customers calling are impatient; hence phone requests have a higher priority over e-mail messages. E-mails that are not responded to within a specified time limit can be prioritized. The goal of this paper is to assess the performance improvement via cross-training the agents. The performance of contact centers operated under different strategies are compared. An extensive simulation study is presented that shows that strategies permitting pre-emptive-resume policies provide the best performance for phone calls. The results also demonstrate that limited cross-training with two skills per agent results in considerable performance improvements. However, the unbalanced traffic intensities due to different mean service times for each class necessitate more cross-training at three skills per agent to have considerable improvement. [Supplemental materials are available for this article. Go to the publisher's online edition of IIE Transactions for the following free supplemental resource: Appendix of additional simulation results]

Server-switching policy for call routing in a multiskill inbound call centre

In this paper, we present a server-switching policy for the routing of calls to Customer Service Representatives (CSRs) in a multiskill inbound call centre utilising Skill-Based Routing (SBR). We have considered two types of scenarios in which a multiskill SBR call centre handling inbound calls can function: (a) with specialist agents only, and (b) with a mix of specialist and flexible agents. We have developed both these models and compared the rewards obtained from each of them. Our models and experimentations help in identifying practices that result in efficient usage of existing personnel in a call centre.

A Quantitative Model for CRM Performance Evaluations

ABSTRACT This paper identifies important performance measurements for the operation of CRM units in the call center industry, and then uses Data Envelopment Analysis (DEA) to compare the performances of a selected set of call centers. The results provide several managerial insights that will assist CRM managers in effective decision making, especially in the areas of increasing efficiencies and improving customer service. Overall, this paper contributes towards managing resources and processes that would aid in the acquisition and support of information technology within a firm. INTRODUCTION Although Customer Relationship Management, popularly referred to as CRM, has been around for almost over a decade, there is little or no consensus about the exact depth and breadth of the CRM concept across a wide array of enterprises. Often time, it is left to the discretion of the managers to define the customer relationship management function to best suit their customer facing interfaces. However, a generic definition outlines the seamless integration of technology and marketing strategy to effectively manage and improve the relationship with existing and potential customers to meet the evolving business objectives. CRM units are used by companies in all aspects of services including financial, healthcare, airlines, recreation, and manufacturing. For instance, IBM's CRM offerings include strategy and change-consulting, system integration, enterprise application implementation for partners such as Siebel, SAP, Peoplesoft, Genesys and Avaya, business reengineering and process redesign, as well as speech recognition and Interactive Voice Response (IVR) (Kelsey, 2002). The early application of CRM was offered as a solution to redefine customer relationships using computer-based tools. However, CRM has evolved both in its functionality and adoption in more recent years. According to a recent Forrester Research estimate, the total CRM market--including applications, infrastructure and services--is likely to grow from $42.8 billion in 2002 to $73.8 billion in 2007 (Prahalad, Krishnan, & Mithas, 2002). But, today's CRM is a dramatically slimmed-down version of the original--it lost its swagger, its acne, and much of the baby fat (Greenberg, 2002). CRM is still expensive and still little cloudy of its benefits; but when it works, it yields great results and remains worthy of the risk of investment. With the IT budget squeeze in recent years, the need to demonstrate the connection between IT spending and business results has become even more critical. Customer satisfaction remains a major objective in CRM operations. Higher customer satisfaction facilitates building and cultivating customer loyalty that results in the ability of a company to retain its existing customers and increase the value of these customers via cross-sell and up-sell (Karbowiak, 2002). Having an IT staff that understands the business and its customers' needs is one key to improving customer satisfaction (Chabrow, 2002). Investing in skilled IT personnel who understand the clients and the business process is probably more beneficial than investing on pure technology. The human factor is very important in implementing an IT system. If the product is too cumbersome to use, employees will quickly find ways to work around it, defeating the very purpose of its implementation. With most IT issues, there is, as the Gartner Group puts it, the Hype Cycle (Fernandes, 2002). The CRM glitter has passed by, and realty has set in. It is time now to aim for operational efficiency. CURRENT LITERATURE Klenke (1995) identifies call center managers as matchmakers who decide which agent can best meet a particular caller's needs, using a skill-based routing system. Bordoloi (2004) offers a quantitative model for determining an optimal recruitment policy amidst vibrant turnover rates in knowledge-intensive call centers. …

Value‐Based Routing and Preference‐Based Routing in Customer Contact Centers

Telephone call centers and their generalizations—customer contact centers—usually handle several types of customer service requests (calls). Since customer service representatives (agents) have different call‐handling abilities and are typically cross‐trained in multiple skills, contact centers exploit skill‐based routing (SBR) to assign calls to appropriate agents, aiming to respond properly as well as promptly. Established agent‐staffing and SBR algorithms ensure that agents have the required call‐handling skills and that call routing is performed so that constraints are met for standard congestion measures, such as the percentage of calls of each type that abandon before starting service and the percentage of answered calls of each type that are delayed more than a specified number of seconds. We propose going beyond traditional congestion measures to focus on the expected value derived from having particular agents handle various calls. Expected value might represent expected revenue or the likelihood of first‐call resolution. Value might also reflect agent call‐handling preferences. We show how value‐based routing (VBR) and preference‐based routing (PBR) can be introduced in the context of an existing SBR framework, based on static‐priority routing using a highly‐structured priority matrix, so that constraints are still met for standard congestion measures. Since an existing SBR framework is used to implement VBR and PBR, it is not necessary to replace the automatic call distributor (ACD). We show how mathematical programming can be used, with established staffing requirements, to find a desirable priority matrix. We select the priority matrix to use during a specified time interval (e.g., 30‐minute period) by maximizing the total expected value over that time interval, subject to staffing constraints.

A multi-class fluid model for a contact center with skill-based routing

A multi-class deterministic fluid model is proposed to describe and improve the performance of a customer contact center with skill-based routing. The fluid model can be regarded as an approximation for a stochastic queueing system with multiple customer classes and multiple server groups, with customer abandonment and non-exponential service-time and time-to-abandon distributions. The fluid model is attractive to provide a rough analysis of large systems, with high arrival rate and many servers. Even though the fluid model evolves deterministically, the service-time distributions and time-to-abandon distributions beyond their means play a critical role. The fluid model can be used for staffing, routing and system design, because it is possible to formulate tractable optimization problems.

A Staffing Algorithm for Call Centers with Skill-Based Routing

Call centers usually handle several types of calls, but it is usually not possible or cost effective to have every agent be able to handle every type of call. Thus, the agents tend to have different skills, in different combinations. In such an environment, it is challenging to route calls effectively and determine the staff requirements. This paper addresses both of these routing and staffing problems by exploiting limited cross-training. Consistent with the literature on flexible manufacturing, we find that minimal flexibility can provide great benefits: Simulation experiments show that when (1) the service-time distribution does not depend on the call type or the agent and (2) each agent has only two skills, in appropriate combinations, the performance is almost as good as when each agent has all skills. We apply this flexibility property to develop an algorithm for both routing and staffing, aiming to minimize the total staff subject to per-class performance constraints. With appropriate flexibility, it suffices to use a suboptimal routing algorithm. Simulation experiments show that the overall procedure can be remarkably effective: The required staff with limited cross-training can be nearly the same as if all agents had all skills. Hence, the overall algorithm is nearly optimal for that scenario.

Service Management and Employment Systems in U.S. and Indian Call Centers

The explosive growth of call centers in India has gained widespread attention because of its potential impact on employment in the United States and other advanced economies. Media accounts report that Indian operations are more likely to use college-educated workers while paying one-tenth of U.S. wages. Some argue that these advantages may allow Indian centers to outcompete U.S. centers on both cost and quality.1 Nonetheless, complaints about poor quality and security, as well as consumer backlash, have led some firms to pull out of India, while leaders in the offshoring business such as General Electric have sold their Indian operations altogether. High turnover rates have become a particularly serious problem in recent years as an expanding number of employers compete for a small pool of educated employees, a trend that both increases costs and undermines service quality. With heated debate more prevalent than systematic empirical investigation, our understanding of this emerging sector is based largely on anecdotal evidence. National figures on employment, industry trends, and the percentage of centers operated in-house (as opposed to outsourced or offshore) are unreliable.2 [End Page 335] Our own national survey of U.S. call centers suggests that after two decades of rapid growth, the outsourced sector represents less than 15 percent of the market; and Indian offshore centers cover a tiny fraction of the U.S. market. In addition, there has been little or no research on management and employment practices in this sector, either in the United States or in India. In this paper, therefore, we consider two questions. First, how similar or different are call center management strategies and employment systems in each country? Here our goal is to map the management practices adopted by three types of operations: in-house centers in the United States, outsourced centers in the United States, and offshore centers that are owned and operated by subcontractors in India and serve the U.S. market. Are there systematic differences in these practices, or is there a call center "production model" that has diffused across very different institutional and organizational contexts? Second, what are the implications of different management practices for outcomes such as turnover? In other words, which practices explain the high levels of turnover in the industry? To answer these questions, we draw on an original establishment-level survey of 330 call centers in the United States and India. We focus on customer contact rather than back-office operations such as check processing or online order fulfillment. For each center, the survey provides information on the customer base, market and ownership conditions, organizational characteristics, work functions, workforce skills and training, call center technology, work organization, compensation, and outcomes such as absenteeism and turnover. In the next section, we discuss prior research that informs our study. We then present the study methods and analytic strategy and our findings. Finally we outline the study's limitations and implications for policy. [End Page 336] Prior Research The first question we address in this study concerns the extent to which call center management practices vary across markets and institutional settings. Call centers represent a new industrial model driven by advances in information technologies that are now ubiquitous. These technologies facilitate the automation of services through interactive voice recognition units, standardize customer transactions through skill-based routing systems, create machine-paced operations through automated call distribution systems, and routinize work through widespread use of scripting and electronic monitoring. However, research shows that service management strategies and employment systems vary substantially across centers that serve different industry and customer segments, and that perform different work functions—from professional approaches to service to highly transactional or cost-driven ones.3 In this line of research, work and employment systems typically are defined to include three dimensions: (a) the level of education and training required...