Multidimensional Assessment of Call Center Customer Service Representative Performance Using Association Analysis and Anomaly Detection MethodsPages 90-102
Abstract:
Call centers are a critical component of customer service and sales‐focused activities, and in outbound calling operations, the performance and behaviors of customer service representatives constitute a key factor in securing competitive advantage for firms. However, traditional evaluation approaches typically concentrate on unidimensional metrics and fail to capture the multifaceted activities and anomaly tendencies of representatives throughout the workday. Against this backdrop, the present study seeks to remedy this shortcoming in current call center operations by examining representative performance within a multidimensional analytical framework, thereby offering a novel perspective both theoretically and practically. To this end, datasets from a telecommunications‐sector call center were consolidated across three separate datamarts, and analyses were conducted at both daily and monthly levels. The methods of clustering, TOPSIS, anomaly detection, and association rule mining were applied in an integrated fashion, enabling the identification of behavioral profiles and nonstandard behaviors among representatives in different segments. The results furnish operations managers with the capability for real‐time intervention in the short term, while yielding targeted insights for employee training and development in the long term. This study advances a multidimensional performance management model that transcends unidimensional metrics by incorporating behavioral data, thereby contributing to both organizational behavior theory and the data science literature. The proposed approach not only supports concrete steps toward workforce optimization, enhanced customer satisfaction, and improved overall service quality within the call center ecosystem but also provides a robust methodological foundation for future research.
Keywords: Call Center, Performance Evaluation, Clustering Analysis, Anomaly Detection, TOPSIS
|
PDF 