The security of supervisory control and data acquisition (SCADA) networks has attracted considerable attention since the discovery of Stuxnet in 2010. Meanwhile, SCADA networks have become increasingly interconnected both locally and remotely. It is, therefore, necessary to develop effective network intrusion detection capabilities. Whitelist-based intrusion detection has become an attractive approach for SCADA networks. However, when analyzing network traffic in SCADA systems, general properties such as TCP handshaking and common ports are insufficient to create flow whitelists. To address the problem, this chapter proposes a methodology for locality-based creation of flow whitelists and conducts experiments to evaluate its effectiveness in seven SCADA systems. The experimental results demonstrate that the methodology generates effective whitelists for deployment in SCADA networks.