Meeting NERC-CIP Requirements

Issue 9 and Volume 112.

Cyber security rules are forcing big changes in networking and computing infrastructure.

By Kevin Staggs, Honeywell Process Solutions

As deadlines for the North American Electric Reliability Council Critical Infrastructure Protection (NERC-CIP) standards approach, power plants and electric utilities must begin implementing the necessary security practices to meet the compliance requirements.

In the process and manufacturing industries, the advent of open architectures and standard protocols presents both new opportunities and risks for plant enterprises. While the evolution from isolated proprietary applications to open technology has expanded business information availability, it has also exposed the enterprise to electronic threats.

Most process control systems used today were not originally designed to defend against cyber attacks. The events of 9/11 and recent cyber incidents on control systems have increased awareness of the inherent vulnerabilities of control systems connected to the Internet and remote telephone connections.

Recently, the Federal Energy Regulatory Commission (FERC) approved eight new mandatory critical infrastructure protection (CIP) reliability standards designed to protect the nation’s bulk power system against potential disruptions from cyber security breaches. The standards were developed by the North American Electric Reliability Corp. (NERC), which FERC designated as the electric reliability organization (ERO).

NERC Standards and Implications

In June 2006, NERC adopted Cyber Security Standards CIP-002 through 009 specifying the minimum requirements needed to ensure the security of the electronic exchange of information for supporting the bulk power system. NERC-CIP identifies standards in key areas designed to protect power plants and all other aspects of electric utility operations and assets. The standards include provisions for identifying critical cyber assets, developing security management controls, training, perimeter and physical security, incident reporting and response planning and recovery plans.

FERC, in its Notice of Proposed Rulemaking (NOPR) released July 20, 2007, has proposed to approve these standards, with directed modifications, as mandatory and enforceable with sanctions and penalties of up to $1 million a day for utilities found to be noncompliant.

According to NERC-CIP implementation rules, companies must be substantially compliant by December 31, 2008, compliant by December 31, 2009 and auditably compliant by December 31, 2010.

Many electric utilities are unsure how to deal with NERC-CIP regulations. Some power companies are struggling to establish a direction or are taking a “wait and see” approach until there is more clarity around the still evolving regulations.

NERC-CIP establishes standards in eight key areas designed to protect not only power plants, but all other aspects of electric utility operations and assets as well. The standard includes provisions for identifying critical cyber assets (section 002), developing security management controls (section 003), implementing training (section 004), identifying and implementing perimeter security (section 005), implementing a physical security program to protect critical cyber assets (section 006), protecting assets and information within the perimeter (section 007), conducting incident reporting and response planning (section 008) and crafting and implementing recovery plans (section 009).

These eight standards address the same areas covered by the NERC 1200 Standard, adopted in 2003, but with some important differences. For example, instead of requiring organizations to identify their critical cyber assets directly, they must now identify their critical assets and then determine their critical cyber assets. (A critical cyber asset must be dial-up accessible or use a routable protocol for communication.)

In many ways, CIP-002 through CIP-009 set a higher security bar. While the standard has not been finalized, deployment of security best practices will help to address the requirements in a phased manner without requiring a one-time, major investment.

Meeting the Requirements

Security improvements should be part of an enterprise-wide risk management program for all process and manufacturing companies, but the challenges are daunting. Moreover, the two groups that must jointly solve this problem—corporate IT personnel and plant operators—traditionally do not work closely together. At the same time, a general lack of awareness of the serious problem slows progress.

Click here to enlarge image

Process and manufacturing companies need a comprehensive security strategy addressing process, people and technology. Honeywell’s strategy includes assessment of all cyber related assets, analysis of risk, design of mitigation strategies and implementation of those strategies and monitoring/management of the entire security strategy, in order to ensure the security of critical assets. It also uses defensive technologies such as patch management, anti-virus software, intrusion detection and security monitoring. All personnel must understand their role in maintaining the security of plant systems.

A successful and effective plant safety and security solution begins with a strategy incorporating security at all levels of the operation. It is important to keep control systems and networks current with security updates, security procedures, documented best practices, regular assessments and testing.

The first step in meeting the NERC-CIP requirements is a security assessment, which involves gathering knowledge about the environment inside and outside of the organization. This includes awareness of electronic threats before they reach the organization, identifying possible regulatory compliance issues, assessing the effectiveness of security and administration tools and processes and manually validating these security concerns using penetration testing methods where it is safe to do so.

Security policy creation and enforcement establish who is authorized to gain access to what information and perform what functions, measures compliance with these policies and procedures and recommends ways to improve compliance.

Deployment includes design and implementation of security measures and responding successfully to vulnerabilities; securing devices, applications and networks against threats before they occur; and taking steps to ensure that information is up-to-date, compliant and restorable. It also involves recovery procedures and tools in the event an attack eludes other security measures.

Finally, security monitoring and management provides real-time, 24/7 monitoring and management of security information resources to prevent disruptions and minimize downtime.

Protecting Your System

A transformation from proprietary to open control systems is underway within the process industry. This trend, coupled with the connectivity between open control systems and enterprise networks, has introduced unprecedented cyber vulnerabilities in control systems.

Without an effective cyber security regimen, the fundamental mission of process control—to ensure safe and reliable operations—can be compromised by an ordinary cyber threat. Therefore, a comprehensive cyber security policy is an essential element of every process control and safety system implementation.

Honeywell advocates a layered approach to seamlessly integrate multiple technologies as the most secure way to protect a plant’s people and assets, incorporating security at all levels from the process control network to the perimeter of the plant. At the core of this sphere of protection is process design, ensuring that processes are controlled by a secure process control network (PCN) extending across the entire plant and business networks.

Layers of protection include managing assets to ensure that the process design continues to function as intended, while protecting the plant from pending incidents with an early indication of failing assets. An integrated approach implements tools and procedures for managing abnormal situations and reducing incidents. When an abnormal situation occurs, alarm management solutions help ensure that operators have the information they need in the context they need it. This allows operators to react to situations quickly and accurately, thereby avoiding or mitigating safety incidents.

Next, properly designed emergency shutdown systems and automated procedures can move a plant to a safe state should an incident escalate beyond the inner layers of this protective sphere. If an incident occurs, fire and gas detection solutions, coupled with rapid location of individuals and a carefully designed emergency response procedure, will help contain the impact.

A layered solution approach also protects the perimeter of the facility using access control, asset tracking, perimeter detection and video surveillance.

Cyber Security

Plants should start with a site vulnerability assessment aimed at creating layers of protection from the control network through the perimeter of the plant. Deep understanding of site vulnerabilities is essential before users can take intelligent action to remove or mitigate associated security risks and establish comprehensive layers of protection.

Management must understand the difference between protecting facilities and data, and protecting processes operated or controlled with information technology. Security measures appropriate for traditional IT data networks could be disastrous if inappropriately applied to process control systems resulting in damage to productivity, capital assets and possibly human life.

An effective cyber security plan should include regular risk and vulnerability assessments, hierarchical networks with access restrictions at each level, high-security model deployed on personal computers and servers, physically separated process control and enterprise networks with limited access points, security hotfix and an antivirus deployment strategy and disaster recovery.

Besides properly managing cyber security, it is important to prepare for the possibility of a disaster that could affect critical data. Even a simple hardware failure could jeopardize critical data. Therefore data recovery plans must be in place and allow for rapid recovery through an automated backup and restore application.

With open technology platforms, control system networks in power plants are exposed to the same security threats facing corporate networks, but with the added safety considerations associated with the process industries. These threats include:

  • Indiscriminant, potentially destructive intrusions, such as viruses and worms
  • Network spoofing and denial of service attacks that have performance impacts and potential safety issues
  • Eavesdropping and password cracking that are threats to confidentiality
  • Malicious threats, such as data tampering, impersonation and packet modification.

Power Plant Implementation

To help explain how the NERC-CIP standards can be met, let’s consider an implementation of a sample power plant system configuration. This sample is based upon the emerging ISA SP99 standard, a cyber security standard for manufacturing and control systems.

The ISA SP99 committee has identified a layered structure for manufacturing and control systems. This structure is shown in Figure 1 on page 74. In this diagram the control system is compartmentalized into functional layers of control with the lowest levels being the process valves, actuators and sensors and the protective system. Moving up to the next level are the controllers connected to the process or Basic Process Control. This level is intended to control the system without human interaction. Above the Basic Process Control level is the Area Supervisor Control level, the first level where the HMI interface is introduced. The next level up is Site Manufacturing Operations where traditionally the entire plant is connected. The next level up is Site Business Planning, the first level where the IT and security management of the system is done by the traditional IT organization. Levels 3 and below comprise the manufacturing and process control system. The attention of the ISA99 committee is focused on these levels. This power plant example will do likewise.

The system at Level 3 and below has a different management philosophy than that of traditional IT systems. Since the primary purpose of a manufacturing and control system is to make products around the clock, availability is a primary security attribute of the control system. This drives a different security management process for the manufacturing and control system. Because of this there is a standard practice to install a firewall between the level 3 and level 4 networks.

High Security Network Architecture will meet the CIP-005 requirements and help with the conformance to CIP-007. CIP-005 is met by segmenting the system into levels as defined by the ISA99 standard. In our example, the network is partitioned into network levels creating security perimeters in the system. This first security perimeter is created with a firewall located between Level 3, Level 4 and the Level 3.5 DMZ. The DMZ provides an additional security zone and also an area for the data servers which move the data between the systems at level 3 and level 4. In this example, an historian is located in the DMZ to achieve data movement and it is possible to provide a view of the process through the DMZ without having to log on to the process control system. An additional security perimeter is added between level 2 and level 3 by providing a set of routers that limit communications between the two levels.

The example provides a configuration which will meet the requirements of CIP-007 for Systems Security Management, including anti virus software and security updates. To meet these requirements there must be servers that provide the updates to the nodes at level 3 and below. A Virus Protection Server provides the updates necessary to the manufacturing and control system below it at level 3. Updates are obtained from update servers located in the company’s IT network. A separate server is required so that the deployment of the updates can be managed by the process control team rather that the corporate IT team.

CIP-009 requires that recovery plans for the process control system be in place. Part of creating recovery plans is to document and test the backup and recovery processes for the control system. Most backup programs are not capable of backing up locked files and most control systems lock critical files as part of their normal operation. Most IT backup systems will take the IT subsystem off-line for a period of time to perform the backup. It is important to ask you control system vendor to provide a backup subsystem which will provide the necessary functionality for backing up a process control system on-line.

One final area in the fictitious power plant is a strategy for meeting the CIP-006 Physical Security Standard. This requirement can not be met directly with the control system but integrating this information into the control system can provide for a higher level of safety at the power plant. The physical security system will be able to:

  • Identify and control who enters and exits the facility
  • Track movements of building occupants and assets
  • Control access to restricted areas
  • Track and locate equipment, products and other resources
  • Track the location of personnel on site in the event of an incident
  • Integrate control and security systems for greater speed and efficiency
  • Respond proactively to alarms and events

There is an advantage to integrating the physical security and process control systems together to provide a common alarms-and-events structure. Process operators may need to know about physical security alerts. For example, if an unauthorized person enters a critical process area the process operator may need to take action to protect the plant or the person. Having the systems integrated allows for the operators to know this. It may also be beneficial for plant security personnel to know about certain types of process or plant upsets.

The impending NERC-CIP standards are forcing significant changes in the utility networking and computing infrastructure. Companies can look at cyber security as merely an intrusive requirement, forcing investment in security-specific technologies. Alternatively, they can view this as an upgrade to their infrastructure that creates opportunity for much broader improvements and benefits.


  1. http://csrp.inl.gov/documents/MigationsForVulnerabilitiesCSNetISA.pdf
  2. Shaw, J., “NERC/CIP Compliance: Headache or Opportunity,” Utility Automation & Engineering, (July 2007)
  3. Lindstrom, J., “Inside the NERC CIP Standards,” (Sept. 2005), http://www.symantec.com/business/library/article.jsp?aid=IN_091105_inside_nerc_cip_standards
  4. http://www.isa.org/MSTemplate.cfm?MicrositeID=988&CommitteeID=6821

Author: Kevin Staggs is a 31-year Honeywell employee with experience in process control systems as a hardware, software and systems engineer. He is a member of the global architecture team in the role of global security architect. In this position, he leads a global design for security initiative to define a consistent security architecture and architectural methodology for security development processes.