Proceedings of the 2015 Winter Simulation Conference
L. Yilmaz, W. K. V. Chan, I. Moon, T. M. K. Roeder, C. Macal, and M. D. Rossetti, eds.
USING SIMULATION TO ASSIST RECRUITMENT IN SEASONALLY DEPENDANT
CONTACT CENTERS
Leeanne May Peer-Olaf Siebers
School of Computer Science School of Computer Science
University of Nottingham
Jubilee Campus
University of Nottingham
Jubilee Campus
Nottingham, NG8 1BB, UK Nottingham, NG8 1BB, UK
ABSTRACT
The weather is unpredictable and can have a large impact on the profitability of seasonal businesses,
particularly if staffing requirements are highly temperature-dependent. In this paper we describe our
efforts in developing a what-if analysis tool to assist affected Small and Medium Enterprises in
determining the best case scenario for timing hiring new staff and deciding the optimum length of
temporary employment contracts. Together with a boiler maintenance company we have developed a
prototype simulation tool that can be employed by users with minimal statistical and modelling
knowledge. Our usability tests with the boiler maintenance company confirmed the usefulness of the
developed tool as a decision support aid for managers. In this paper we focus on describing the tool
development and testing process. With regards to real world experimentation we are still waiting for the
feedback from the company.
1 INTRODUCTION
Payroll costs to companies can make up anywhere from 18-50% of operating expenses (SHRM 2008).
For businesses which must substantially increase their full time equivalent (FTE) members of staff during
peak periods, it is important to get the timing of recruitment correct to avoid unnecessary costs. When the
peak period is decided by something as variable as the weather, this decision is particularly difficult.
There has been a lot of work carried out in the area of Contact Centers (CC) and hospitals to
determine the optimum number of staff needed based on the service requirements of the business (e.g.
Ernst et al. 2004). Less research has been conducted into what period of time is needed to achieve the
staffing levels required. For example, it may take over a month to recruit and train a new employee, and
over seven weeks before they are fully trained (CFA 2012). If recruitment starts too early then staff will
have increased idle time; recruit too late and the work increases faster than staff can cope, leading to
complaints and lost customers. Since some businesses hire up to three times as many FTE at peak times,
this can cause significant damage to business if misjudged (Myron 2002).
Energy companies, for instance, are heavily affected by the weather. Temperature is the main driver
of gas use in the UK (CelsiusPro 2015). This means that the changes in temperature can be a huge strain
on boilers and thus boiler repair companies. Academic work to date, which includes ambient temperature
measurement of "degree-days" as a way to forecast consumption (Quayle and Diaz 1980) has focused on
energy companies. In our study we work with a company that provides boiler maintenance and repair
services. As gas-based central heating systems are the primary way energy customers heat their homes,
we can apply the same principles to a boiler repair and service company (Meier and Rehdanz 2010).
1573 978-1-4673-9743-8/15/$31.00 ©2015 IEEE