Can flexibility be constraining?

Abstract

Five common options for workforce flexibility and their robustness under uncertain demand are investigated. In the first stage, a firm makes optimal staffing decisions according to estimated demand and a given workforce flexibility policy. In the second stage, it reallocates its workforce to react to demand shocks. Numerical results are presented that show that flexibility can lead a firm to staff with too little slack to be flexible to demand shocks, thus leading to higher total costs, i.e., staffing and inventory costs. The forms of flexibility that give robust benefits are identified and an analysis on how different forms of flexibility interact with each other is performed.

[Supplemental materials are available for this article. Go to the publisher's online edition of IIE Transactions for the following supplemental resource: Appendix with additional tables of results.]

Keywords:

• Flexibility
• workforce management
• optimization
• scheduling
• cross-training

Notes

¹Depending on the demand pattern, having four starting times is not necessarily more effective than having three starting times whereas doubling the number of start times with a similar distribution within the day is increasing flexibility.

²Although other cross-effects, such as the number of starting times and the part-time ability, may affect the system, we only focus on which flexibility type is more effective instead of which “combination” of flexibility types is more effective. Thus, we discard all other cross-effects.

³We acknowledge this threshold is somewhat arbitrary but it forms a natural break for our results.