Browse Prior Art Database

A Method for Optimizing Staffing Levels in a Ticketing Service System using Simulation

IP.com Disclosure Number: IPCOM000243084D
Publication Date: 2015-Sep-14
Document File: 6 page(s) / 117K

Publishing Venue

The IP.com Prior Art Database

Abstract

A method for optimizing staffing levels in a ticketing service system using simulation is disclosed. The method uses a simulator to find optimal solution in a ticketing service system.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 36% of the total text.

Page 01 of 6

A Method for Optimizing Staffing Levels in a Ticketing Service System using Simulation

Disclosed is a method for optimizing staffing levels in a ticketing service system using simulation. A ticketing service system may be considered in which incoming tickets are classified based on service complexities. For each service complexity, a pool of resources with the proper skills to serve the tickets may be used. A resource from the pool of resources with skills in a given complexity class of service may also serve other classes of service with lower complexity. A queueing model may be used for allocating a queue for each service complexity class. The queueing model parameters correspond to one of a ticket arrival process, a ticket service time process and pool operations, wherein the pool operations consist of dispatching policies and swinging resources between the various queues, shift schedules and official breaks for the resources. Queues may not be analyzed independently of each other because the pool dispatcher may temporarily move a resource from one queue to serve another queue with lower complexity of service if the length of the latter queue reaches a certain threshold.

An optimization mechanism may be used to determine the minimal number of pool resources required for each skill level per shift schedule. The minimum number of pool resources is determined to achieve a response time Service Level Agreements (SLA) associated with the incoming tickets. Simulation may be used to obtain accurate estimates.

In an embodiment, the optimization mechanism may be used for fixing a single service class. The optimization mechanism may fix the single service class by determining a lower bound for the optimal number of resources to be allocated for the single service class for each shift schedule. The lower bound for the optimal number of resources is determined based on ticket arrival rate, ticket service rate and pool resources work schedule for the service class. Further, the optimization mechanism is used for developing a process that efficiently searches for an optimal allocation per shift starting from the lower bounds and iteratively invokes a simulator.

In another embodiment, the optimal solution obtained for each single service class by the optimization mechanism is used and effects of shifting resources from higher level service class queues to lower level service class queues are considered. Thereafter, a search mechanism is developed that uses individual optimal allocations for each service class obtained and iteratively invokes the simulator. The search mechanism is developed to find an optimal solution when the tickets of all service classes are combined into one single stream.

For each service class, a necessary condition for system stability is that the average arrival rate of tickets should not exceed the system capacity to serve the tickets. The stability equation for this system is given by:

1


Page 02 of 6

N

i i


i...