Control of contact center model functional parameters to reduce the load on agents

Abstract

The operation of the contact center model has been reviewed, and it has been determined that one of the center’s key objectives is to maintain the minimally required number of agents without call interruptions. For further analysis and the development of a mathematical model for managing a networked contact center, we have identified the key performance indicators (KPIs) for agent efficiency, as well as the core metrics for assessing overall contact center performance. From a technical perspective, a typical contact center is an integrated hardware-software system. Services are executed on servers running specialized software integrated with a CRM system.

The contact center is conceptualized as a dynamic system composed of multiple interacting subsystems. A mathematical model for its operational control integrates statistical data, system feedback loops, and evolving management techniques to enable resource optimization. This model considers: matrix Bi,j - a representation of controllable parameters for each managed object (e.g. operator profiles, call attributes), where each bij is a random variable with known distribution; time delay decomposition total latency Ti , capturing queueing delay, transfer time, and computational processing per cycle. Has been responsed time calculations based on formulas that integrate equipment utilization K, queue depth, and probabilistic response models Bernoulli-type service distributions.

By incorporating has been proposed mathematic model and tical model for calculating the core probabilistic-temporal performance metrics – supplemented by numerical examples – the proposed framework can extend the functionality of the call processing module CPM without complex configurations.  It is obtained that load thresholds K > 0.8 activate redistributive mechanisms to balance operator workloads or route calls to backup services. By analyzing traffic intensity, queue dynamics, and call handling performance across service cycles, the model supports adaptive adjustments to operator assignments and system capacity in near real-time. It also contributes to the formation of Key Performance Indicator (KPI) frameworks without requiring complex reconfiguration, enabling flexible enhancements to call routing modules.