Browse Prior Art Database

A System and Method for Integrated Capacity Planning

IP.com Disclosure Number: IPCOM000217992D
Publication Date: 2012-May-15
Document File: 5 page(s) / 86K

Publishing Venue

The IP.com Prior Art Database

Abstract

Disclosed is a decision support system that forecasts the incoming call volume at a call center and recommends an optimal seating requirement based on a multi-criteria scheduling of the employees that takes into account their varying proficiencies. Also described is a user interface that is used to systematically display the recommendations.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 38% of the total text.

Page 01 of 5

A System and Method for Integrated Capacity Planning

1 Features


1.1 Forecasting Mechanism

Generate forecast based on historical volumes incorporating relevant external events such as new launch of products, announcements, change in terms, etc.


Reconcile inputs from different organization's areas , reports, and data points.


1.2 Decision Support System

Scheduling based on proficiency of the agents i.e. capability to fulfill demand based on variable skills/proficiency (inversely proportional to average handling time for calls) of the workforce.


Recommend optimal seats requirement for capacity planning managers for every account.


Method to generate recommendations on seat requirements, by leveraging cross lines of

business at an enterprise level.


1.3 Capability to handle accounts with different SLA requirements


Create schedules to meet SLA maximizing the efficiency of resources.


1.4 Design of a utility function


Address problems like training, head count planning, redeployment, seat utilization, login times via a composite utility function.


1.5 Design of the head count planning function


Method to generate head count plans, leveraging information about agent redeployment across various lines of business.


1.6 Design of a user interface


User interactive capability to get business inputs, reconcile them to lock forecasts for

planning.


Display information about future demand in a meaningful manner. Provide information about the shape of the demand curve to the end user, so that the user may appropriately modify the demand curve, across various months, as per business intelligence.

1


Page 02 of 5

Figure 1: Schematic description of the system


2 Technical Aspects and Approach


2.1 Demand Forecasting Method


Forecasting Historical Call volumes are taken and ARIMA(p,d,q) model is used to predict the call volume
on a daily basis. Values of (p,d,q) are estimated based on statistical analysis and is data dependent
hence timely calibration is required.

Reconciliation Users provide input for both shape and lift of the curve based on the industry scenario,
sales trend, some anticipated new deals. Base forecast is adjusted by weighted mean method based
on their inputs where weights are calibrated based on historical inputs and their accuracy using an
algorithm.

Demand Generation Demand for each account is generated based on the locked forecasts for the planning
horizon. Forecast is published to the workforce management team for further resource and capacity

planning.


2.2 Scheduling

Resources can belong to different classes where each class has parameters like productivity and

proficiency.

2


Page 03 of 5

Proficiency is defined as the amount of time an agent takes to deliver the unit amount of service. Productivity
is defined as the amount of productive hours out of total working hours. Resources move up the value chain
by acquiring new skills and improving their proficiency. We model the proficiency curve as a

piece wise linear
function. A snap shot of supply poo...