In May, the pilfering of 500 feet of copper wire from a subway line in New York City left hundreds of travelers stranded and caused millions of dollars in damage. And this wasn’t an isolated incident. There have been instances of copper theft on Seattle's Sound Transit, Long Island Railroad, and San Francisco’s BART system. In fact, the National Insurance Crime Bureau saw 33,775 insurance claims for metal thefts from 2010 to 2012, and 96 percent of those were for copper alone.
Transit systems, which utilize copper in wiring, are increasingly targeted by thieves looking to make some easy cash. Although $2.50 for a pound of copper might not sound like a lot of money, when you have hundreds of feet of copper wire, you’re talking about thousands of dollars or more. The thieves might make off with up to $100,000 worth of copper, but it might cost the transit authority $2 million to fix the damage, to say nothing of the operational impact of a rail system that has been completely shut down.
This is the paradox of copper theft: While the thefts are costly and damaging to the railway infrastructure, the real cost impact is operational. As the industry struggles to get a handle on copper theft, it may see a bigger payback by addressing the operational impact in addition to protecting infrastructure. It’s not just about preventing the theft — it’s also getting the trains back on a normal schedule when incidents like this take place.
The Challenges of Preventing and Mitigating Theft
The first step to defending railway infrastructure is to ensure the proper security systems are in place – for instance, a video monitoring or intrusion detection system. Cameras, which can help identify thieves and intruders, can also provide incident management support to track how the individuals got into the area and exactly what they did while they were there. In particular, infrared cameras are extremely useful for outdoor surveillance since they can record at any time of the day or night with clarity, though they are more expensive.
Intrusion detection systems trigger an alarm and register an alert when someone trespasses or breaches the perimeter of fenced in rail lines, depots, or stations. These systems can also call to focus any available video feeds of the breached area. But not all rail lines are protected by a fence, and thieves are more likely to strike in remote, unprotected areas. In these circumstances, a transit operator may not know a copper theft has occurred until it’s too late and the damage is already done.
Challenge: Transit systems are large. There can be dozens of rail lines spanning hundreds, even thousands of miles of track across a broad geographic area, making comprehensive and continuous surveillance of the railway nearly impossible. The New York City subway system has 656 miles of track, 40 percent of which is above ground. And outside the U.S., even small countries can easily exceed 5,000 miles of track. This is a globally universal challenge.
Putting intrusion detection systems in place in enclosed areas such as depots or stations is relatively straightforward, but protecting right of way using such systems is cost-prohibitive. Placing a video camera every 50 yards in the New York City subway, for example, would require almost 25,000 cameras. It’s clear that there are some parts of the railway network that cannot be protected through direct visual monitoring.
Challenge: Reliably detecting compromises in transit systems. Even if it is impossible to watch every inch of the track, transit systems should have a way of monitoring the operational status of its track in real time. That’s why transit systems should also ensure the tracks and other facilities use sensors that can detect and alert the command center when power to the third rail has been cut or copper pulled from the track.
However, here is a case where the architecture of the rail network can become a liability. In the case of the New York theft:
Sergeant Kevin Cooper, a veteran investigator on the case, said trains didn’t stall immediately because redundancy in the power systems allows them to keep moving for possibly another day after copper is removed.
In actuality, the lack of power to the third rail wouldn’t have been noticed until a train attempted to use the track (and thereby complete the circuit). While a fully operational train is the ultimate test of the serviceability of the track, it shouldn’t be the first and only test. Copper theft is not a smash-and-grab crime. Removing 500 feet of copper wire isn’t something that took the perpetrators only five minutes. If a railway has a way of detecting the change and acting on it in real time, there might be a chance to stop the thieves before they cause real damage.
Challenge: The task of monitoring is overwhelming. Even when proper monitoring systems are installed -- video cameras and asset monitoring, for example -- sifting through the many alerts they generate to find those that indicate a security threat is a challenging undertaking. Who is going to watch the 70,000 video cameras, 24 hours a day, seven days a week?
In the case of New York, there may have indeed been an alert that the tracks had switched over to the backup power system, but this alert might easily have been buried in the hundreds of other routine alerts that were being generated throughout the transit system. The signal-to-noise ratio can be a critical factor in whether the people monitoring a situation will pick up on an important alert.
Rapid Situational Awareness is Key
Rapid situational awareness and communication to law enforcement could help mitigate copper theft and other situations in real time. In any failure situation, the dissemination of information is critical to improving response time. One of the main bottlenecks is the time it takes to correlate inputs from multiple disparate data sources into meaningful situation data. Railway operators must address the disparity between information required and information available.
Immediately after a crisis occurs, the demand for information is extremely high. However, there is a big disparity in the data available, because hierarchical communication topologies typically used for the collection and dissemination of data are too slow. Railway infrastructure managers must coordinate with a range of law enforcement and emergency agencies (from police and fire control to ambulance and homeland security), as well as external companies (such as towing services and handlers of hazardous materials) in order to get the situation under control. Today most of these exchanges occur by telephone or radio, which means that the various stakeholders need to rely on verbal cues. There is no common operating picture or shared response plan.
Integrate and Automate with a PSIM System
Transit agencies should consider deploying a physical security information management (PSIM) system that collects and analyzes data from video cameras, intrusion detection systems, GIS, track power circuit sensors, and other sensors. With a PSIM, command centers can view this information in real time and receive immediate alerts to incidents such as potential copper thefts. Security personnel can then provide information about the alert -- such as its specific location and the nature of the breach or failure -- to first responders so they can respond quickly and appropriately.
PSIMs address the overwhelming amount of information coming in from video monitors and other sensors. The remote devices generate an alert anytime something abnormal is happening, though the nature of these things is that they report both minor and major abnormalities with an equal sense of urgency. PSIMs cut through this noise to surface the critical events, ensuring that when something goes wrong the right people are doing the right thing at the right time. The idea is to increase the availability of information as the crisis is breaking (again, Figure 1).
Have a Plan and a Process
The second crucial step to infrastructure security is to have a plan and process in place to dictate which alerts should be responded to first so there aren’t any oversights. After all, it’s not only imperative that the right people are alerted to the issue, but that an action is performed to correct the situation.
There’s a detection problem, and then there’s the “doing the right thing quickly” problem. There are many instances where incidents have been detected but the right steps weren’t taken. PSIM systems automate the prioritization of alerts and tasks and provide approved procedures for remediating issues, so if a monitor indicates a copper theft in progress, security personnel can respond in real time. A task list will prompt personnel on what exactly needs to be done, and by who.
The Real Issue is the Cost of the Downtime
Not to downplay the costs of copper theft, but the biggest cost in events like this that cause a disruption of service is the disruption itself.
A UK National Audit Office (NAO) report from 2008 (source: www.nao.org.uk) provides valuable insight on the impact of failures on train delays. The report shows that failures account for 40,969 incidents and 3,040,686 minutes of delays (out of a total of 14,000,373 minutes of delays per year), costing the UK economy at least $US 330 million (€300 million) per year.
The final conclusion of this report was that the majority of the delays stemmed not from the infrastructure itself but rather from inefficient incident handling. This can be seen in the following chain of events:
• The Wimbledon Area Signaling Center noted a track circuit failure - which was probably the result of damage caused during overnight engineering work — at 4:45 a.m.
• The Signaling and Telecommunications team already in the area began to diagnose and repair the fault.
• The local Mobile Operations Manager was sent to the site at 7:10 a.m.
• A command structure was set up on site at 8 a.m while repairs were made to a number of damaged parts, which were completed by 10 a.m.
Total delay minutes attributed to the incident: 1,868. Estimated value to passengers: £200,000 ($313,000).
It’s not about stopping copper theft, because even if a transit agency does that, there are 40,968 other things that go wrong every year. The real question is, if the system knows that something bad is happening, how does the railway infrastructure organization prioritize its actions to address the right anomalies and failures to minimize the impact on passengers? That’s the real issue facing transit agencies, of which this copper theft is just one small part.
Rail infrastructure agencies -- whether it’s BART, NJT, MTA or MetroRail — have a holistic responsibility on the rail infrastructure and services. They are expected to meet a specific KPI in terms of punctuality and service delivery, and, if they don’t, it doesn’t really matter why — bad weather, poor infrastructure, or criminal activity. This is why they need a holistic solution, because they’re measured on the holistic performance of the system.
Udi Segall is a director of Business Development with Qognify.
