The Edmonton Folk Festival (www.efmf.ab.ca) is a four-day outdoor event that has been held annually since 1980. Every year thousands of people attend afternoon workshops and evening main-stage performances featuring local artists and internationally renowned musicians. The success and continuity of the Folk Fest is dependent on the ongoing dedication of volunteers.

We had the opportunity to work with two gate crew coordinators from the Edmonton Folk Festival to develop an automated scheduling tool that incorporated the unique features and individual preferences related to scheduling volunteers. The tool was used to schedule several dozen volunteers for the 2003 and 2004 Folk Fests.

Employee scheduling is a common O.R. problem. Unfortunately, employee scheduling is typically situation-specific. Methods used to schedule airline, medical and retail personnel do not fully apply to scheduling volunteers for annual events. Volunteers offer their services to sporting and cultural events for a variety of reasons such as interest in the activity, to gain experience and for social interaction. But unlike paid employees, volunteers choose to continue to serve based on intangibles such as their enjoyment of the crew they work with, their satisfaction with the shifts they are assigned and whether they feel valued. If these criteria are not met, a volunteer may walk away.

Prior to 2003, the gate crew coordinators at the Folk Fest manually drew up the schedule. The process was time-consuming, frustrating and susceptible to errors. The festival had no funds to buy scheduling software, and even if it had, typical software programs don't address the many constraints and preferences unique to volunteer scheduling. The problems of scheduling volunteers for a four-day event include limited availability of volunteers, the acyclic nature of the problem, atypical constraints and the high priority placed on satisfying volunteer preferences. The coordinators required an Excel-based decision-support tool that would allow them to generate shifts, input volunteer information and preferences, schedule volunteers by maximizing preferences, and produce accurate schedules and reporting forms.

Employee scheduling models assisted with the first stage of the process. Constraints included ensuring that the schedule met the demand requirements for every time period, each person worked exactly 20 hours, and no one worked back-to-back shifts or late-night/early-morning combinations. In addition, we included options allowing coordinators to designate guaranteed shifts to volunteers and to input volunteer unavailability.

From Implicit to Explicit


One of the challenges throughout the project was to make the implicit rules explicit and to represent them appropriately in the model. This became apparent after reviewing our initial mock schedules. Through our discussions with the coordinators, they revealed a number of implicit rules including the need to distribute volunteers with first-aid training, account for hours worked prior to the start of the festival, override back-to-back constraints according to volunteer preferences, balance a volunteer's shifts between the two gates as well as day and evening shifts, and ensure a minimum number of experienced people on each shift where possible. We continued to refine constraints during the project to better reflect the coordinators' implicit rules in scheduling and to account for other possibilities they realized the model could handle.

Our understanding of the situation and the development of the model was not yet complete even after incorporating the coordinators' implicit scheduling rules. Once the coordinators received a list of volunteers, each person was individually contacted and welcomed to the gate crew. In conversation, many of the volunteers made requests and indicated preferences such as unavailability for some shifts, preference for working double shifts, working with (or not with) a specific person, preferring specific shifts, or indicating physical limitations that would affect their shift assignments. The coordinators' official policy was that crew members had to be available for all shifts. However, as in previous years, the coordinators took note of all the requests and said they would try to accommodate them. But in doing so they realized they were sending mixed messages. In the past, they inevitably got complaints about unfair treatment and became concerned that the volunteers would not return the following year.

Whether to solicit and accommodate preferences was an ongoing source of tension for the coordinators. On the one hand, they feared that not accommodating every request would anger the volunteers. On the other hand, if they maintained their current policy, they worried that volunteers would not feel valued and that the coordinators were setting a double standard when they did accommodate some people. We agreed that they would probably not be able to fill all requests, but we could develop a schedule that would accommodate as many preferences as possible. The final decision was to see to what extent we could accommodate the preferences submitted.

Preference Point System


We used a point system to handle volunteer preferences. In 2003, we looked at the unsolicited preferences indicated by volunteers and assigned points at the coordinators' discretion. In 2004, the coordinators solicited preferences, and volunteers were given points that they could strategically distribute to shifts they preferred. The preference system used is flexible enough to allow volunteers to assign negative points for undesired shifts such as the Saturday night shifts when a feature artist appears on main-stage. The objective of the model we developed was to maximize the number of volunteer preference points.

Although we felt comfortable that we had addressed all constraints and issues related to assigning volunteers to shifts, a suitable set of shifts had yet to be determined. Trying to address anticipated preferences and constraints, we learned that the preferred length of shift differed among volunteers, the shifts had to accommodate set mealtime hours for volunteers, and, since every volunteer was required to do 20 hours, coordinators had no flexibility to make last-minute changes to address overstaffing during slow periods and understaffing at peak times.

Based on past experience, the coordinators assumed that shifts should be four hours in length. When they attempted to create a manual set of shifts with anticipated gate arrivals (i.e., demand), overstaffing and understaffing occurred because of the regimented shift length that needed to fit around meals and demand. While over- and understaffing are issues in for-profit businesses, they are also detrimental to volunteer organizations. During overstaffing the volunteers feel they are wasting their time, and during understaffing they feel stressed and unfairly treated. We were able to develop an optimization tool that determined two-, three- and four-hour shifts that worked around the meal hours and addressed overstaffing and understaffing problems associated with longer shifts. In 2004, volunteers were asked to identify preferred shift lengths and, in addition to three- and four-hour shifts, some preferred five-hour shifts. The coordinators incorporated shift lengths of three to five hours to accommodate these preferences in year two of the project.

Once coordinators entered the information for individual volunteers, incorporated all preferences and selected a shift pattern, the final task was to create a final schedule that the coordinators could distribute to the volunteers. In the previous years, the coordinators had created the schedule manually using word-processing software.

We improved on this process and produced three schedules that met their needs for error checking and readability. The first was a master schedule. The schedule provided a table of all volunteers and all shifts. Using the master schedule, the coordinators could view who was and who wasn't assigned to the different shifts and whether anyone had been scheduled for back-to-back shifts. They could also look across at each person's assigned set of shifts quickly to assess night shifts and gate balance.

We developed a second version of the schedule including all shifts and listed the volunteers working each shift. It assured the coordinators that they had assigned an appropriate number of people to each shift. Third, we produced individual schedules for each volunteer to eliminate errors by volunteers who failed to locate their names under all shifts. Although producing the three schedules was one of the simplest components of the project, the coordinators and volunteers valued them highly. We could create and print error-free schedules in seconds rather than hours, which made last-minute changes possible.

In the final draft of the 2003 schedule, each volunteer had a balance between the two gates and between day and evening shifts. No one had back-to-back or late-night/early-morning combinations unless he or she requested it. Volunteers with first-aid training were spread throughout the shifts. Most important, we incorporated all articulated volunteer preferences. In 2004, when preferences were solicited, all the constraints were addressed and, although many more preferences were submitted, we were able to accommodate almost all of them. Requesting and attempting to accommodate preferences from all the volunteers proved to be a valuable way to improve morale and demonstrate the importance of volunteers to the success of the Folk Fest.

Although we were pleased with the end result, the coordinators wanted a product that they could use independently so that they could accommodate last-minute changes. Working with one of the coordinators, we developed a set of 11 Excel worksheets and a 10-page user manual. The coordinators entered data on volunteers including partners and individuals, first-aid training, required hours, availability for each shift, guaranteed shifts, personal preferences and overrides on standard constraints (e.g., back-to-back shifts). Coordinators were able to run the optimization program and/or make manual changes, and were made aware of any non-default values, violated constraints and unrealized volunteer preferences.

In 2003, the coordinators used the tool we designed to develop the shifts and the schedules. The project was successful because we combined optimization tools and what-if features on a spreadsheet, and we cooperated closely with the end-users. At the end of June 2003, the coordinators felt comfortable using the scheduler. In 2004, changes to the shifts required revisions to the constraints used. Once these changes were made and an optimized set of shift assignments was generated, the coordinators were able to make manual changes and generate a final draft of the schedule.

While O.R. has made a substantial difference to improving the efficiency and cost effectiveness of employee scheduling, in the case of volunteer scheduling, it has also made a contribution to improving a volunteer's sense of enjoyment, satisfaction and worth.



Lynn Gordon is an associate professor at the Faculty of Education and an MBA student at the University of Alberta.

Erhan Erkut is a professor in the School of Business at the University of Alberta. The article describes a pro bono project conducted by the Centre for Excellence in Operations at the University of Alberta School of Business (www.bus.ualberta.ca/ceo). A detailed account of the project, including the mathematical models used, will appear in Interfaces.

Table of Contents

OR/MS Today Home Page


OR/MS Today copyright © 2005 by the Institute for Operations Research and the Management Sciences. All rights reserved.


Lionheart Publishing, Inc.
506 Roswell Rd., Suite 220, Marietta, GA 30060 USA
Phone: 770-431-0867 | Fax: 770-432-6969
E-mail: lpi@lionhrtpub.com
URL: http://www.lionhrtpub.com


Web Site © Copyright 2005 by Lionheart Publishing, Inc. All rights reserved.