Expert Commentary

Amtrak safety, rail transit and Positive Train Control: Research roundup

An updated collection of recent reports, research and analysis relating to Amtrak, rail transit and Positive Train Control.

Amtrak train

In December 2017, an Amtrak train derailed after traveling too fast on a curve in Washington, killing or injuring dozens of people. In May 2015, a train derailed under similar circumstances in Philadelphia, killing eight and sending 185 others to area hospitals.

Investigators are still trying to figure out what happened in Washington. But the National Transportation Safety Board (NTSB) learned that the engineer on the Philadelphia train was distracted by another train and, according to a report the agency released in 2016, the “most likely reason the engineer failed to slow for the curve was … because of his loss of situational awareness.”

The December crash has raised new questions about whether Positive Train Control (PTC) — technology designed to slow trains nearing dangerous conditions — would have prevented these disasters and whether the federal government erred by extending the deadline for railroads to install it.

As the U.S. Government Accountability Office noted in a December 2013 report, PTC was supposed to be implemented by the end of 2015, precisely to prevent accidents caused by human factors. This system and its accompanying 2015 deadline were mandated under the Rail Safety Improvement Act of 2008, which was passed in response to several fatal rail accidents between 2002 and 2008. PTC was described as a “groundbreaking” wireless communications system comprised of “integrated technologies capable of preventing collisions, over-speed derailments and unintended train movements.”

The Federal Railroad Administration (FRA), charged with overseeing PTC’s development and roll out, had not implemented it on the stretch of rail going through Philadelphia prior to the crash there, as  of Vox first noted. News reports indicate that while the technology has been installed on tracks in Washington, it is not yet operational.

In September 2015, the federal government released a report saying nearly 70 percent of railroads were not going to implement PTC until after the Dec. 31, 2015 deadline — an estimated one to five years afterward. Congress extended the deadline to Dec. 31, 2018, although railroads may qualify for more time if they meet certain criteria.

Below is a selection of research and reports relating to Amtrak, rail transit and Positive Train Control that can help inform reporting on these issues:



“Technical and Safety Evaluation of the Southern California Regional Rail Authority Positive Train Control Deployment Project: Challenges and Lessons Learned”
Placencia, Greg; Franklin, John; Moore, James E. Report sponsored by the Federal Transit Administration, Report No. 0112, July 2017.

Abstract: “Positive Train Control (PTC), often referred to as Communication Based Train Control (CBTC), has been on the National Transportation Safety Board’s (NTSB) ‘Most Wanted List of Transportation Safety Improvements’ for several decades as a safety-enabling system. The Rail Safety Improvement Act of 2008 mandated its implementation after the September 12, 2008, Chatsworth, California, collision between trains from the Southern California Regional Rail Authority (SCRRA or Metrolink) and Union Pacific. SCRRA has undergone substantial challenges to integrate PTC into its operations. This report investigates the multilevel challenges—technological, human, organizational, and systematic—that SCRRA faced implementing the new technology as well as many of the lessons the railroad industry can learn from these challenges. Technology alone cannot ensure safety, but a properly-implemented PTC system can develop and promote high reliability practices that enable safe operations throughout an organization. The report examines interactions among the numerous Systems of Systems for their impact on successful PTC implementation.”


“Amtrak Five Year Service Line Plans: Fiscal Years 2017-2021”
National Railroad Passenger Corporation, 2017.

Summary: This 140-page report offers a broad overview of Amtrak operations, including ridership, ticket revenue, financial forecasts and Amtrak’s five-year capital plan. The report briefly mentions Positive Train Control. According to the report: “After six straight years of annual ridership exceeding 30 million passengers, Amtrak’s business is the strongest it has ever been in its 46-year history. Each day more than 20,000 employees nationwide commit to providing superior customer service. In FY2016 we achieved record revenues of $3.2 billion, proving people are recognizing Amtrak is simply the smarter way to travel.”


“A New Alignment: Strengthening America’s Commitment to Passenger Rail”
Robert Puentes; Adie Tomer; Joseph Kan. Metropolitan Policy Program, Brookings Institution, March 2013.

Excerpt: “Amtrak experienced a significant increase in national ridership after 1997. Using Amtrak’s fiscal period of October to September, Amtrak’s total boardings and alightings jumped 55.1 percent from 1997 to 2012. To put this increase in perspective, it outstrips population growth (17.1 percent) more than threefold over the same period and exceeds the growth in real gross domestic product (37.2 percent). With Amtrak setting ridership records for nine of the past ten years, including the new all-time high in 2012, there is a great chance Amtrak’s passenger growth will continue to far outpace growth in population and GDP. In addition, Amtrak’s passenger growth also exceeds all other domestic transportation modes. The most appropriate modal comparison is domestic aviation, since Amtrak and major airlines compete along certain corridors. In this case, Amtrak more than doubled the growth in domestic aviation passengers (20.0 percent) over the same 16-year period. Similarly, Amtrak also far exceeded the growth in driving (measured by vehicle miles traveled per year; 16.5 percent) and transit trips (26.4 percent). All three modes do carry larger aggregate quantities of people, but these growth trends serve as evidence of changing attitudes toward train travel.

In addition to route length, having a direct connection between major metropolitan areas is another driver of higher Amtrak ridership. Across the past 15 years, a consistent group of 10 corridors, all less than 400 miles long, generate around 70 percent of total system ridership. Each of these routes involves many of the country’s 100 largest metropolitan areas and benefit from the higher job and population densities present in those metropolitan cores. The Northeast Corridor is particularly notable in this respect, connected by the metropolitan anchors of Boston, New York, Philadelphia and Washington. These four metropolitan areas house over 35 million people, generate $2.3 trillion in annual output, and share historic and modern relationships. Similarly, all four metro areas suffer from high traffic volumes between them as well as the country’s most congested airspace (New York-Philadelphia), making the rails an attractive alternative to some of the country’s most delayed airports. Indeed, Amtrak boasts 75 percent of the share of the passenger rail/aviation market between New York and Washington.”


“State of the Northeast Corridor Region Transportation System”
Northeast Corridor Infrastructure and Operations Advisory Commission, Summary Report, February 2014.

Excerpt: “Available capacity on the highway, rail, and aviation networks is limited such that all three modes experience serious congestion levels with negative consequences for productivity and quality of life. Aging infrastructure, especially on the highway and rail networks, threatens to reduce the capacity we enjoy today. Existing plans and identified funding sources fail to fully address the capital needs for bringing our transportation system into a state of good repair or building new infrastructure to support growth in the economy. Despite these challenges, advances in technology and new types of intermodal and interjurisdictional coordination offer opportunities for modernizing our transportation system.”


“Rail Safety: Improved Human Capital Planning Could Address Emerging Safety Oversight Challenges”
U.S. Government Accountability Office, GAO-14-85, December 2013.

Excerpt: “[The Federal Railroad Administration] (FRA) has developed a risk-based approach to direct its inspection efforts, but the agency has been slow to implement broader risk reduction planning. FRA has two tools to help direct its inspection efforts — the National Inspection Plan (NIP) and the Staffing Allocation Model (SAM). The NIP process uses past accident and other data to target FRA’s inspection activities, and the SAM estimates the best allocation of the different types of inspectors across FRA regions in order to minimize damage and casualties from rail accidents. However, all eight FRA regional administrators expressed concerns about FRA’s staffing process that relies primarily on the SAM to predict appropriate regional inspector needs, and that does not allow the flexibility needed to accommodate the regions’ changing resource needs. In addition, the Railroad Safety Improvement Act of 2008 mandated safety risk reduction plans primarily for large freight and passenger railroads. FRA has not yet issued the final rule directing railroads to develop the plans, which was mandated to be issued by October 2012. According to FRA, the rulemaking was delayed due to concerns by railroads over their potential liability. Although FRA anticipates completing approval of railroad’s plans by 2016, the agency has not developed an interim plan with specific timeframes to ensure that there are no further delays in issuing regulations and that timely evaluation and approval of the railroads’ risk reduction plans occurs.”


“Rail Safety: Preliminary Observations on Federal Rail Safety Oversight and Positive Train Control Implementation”
Susan A. Fleming. U.S. Government Accountability Office, Testimony before the Committee on Commerce, Science, and Transportation, U.S. Senate, June 2013.

Excerpt: “According to FRA officials, in the next 5 years, about 32 percent of FRA inspectors will be eligible to retire. Although FRA officials said that they anticipate being able to replace inspectors, it can take 1 to 2 years to find, hire, train, and certify a new inspector. Finally, FRA faces other ongoing and emerging safety challenges like addressing adverse weather conditions and their impact on railroad operations and equipment, educating the public on the potential hazards of rail-highway crossings, accommodating changes in rail safety risks including new freight flows that affect the need for inspections, and hiring and training a specialized inspector workforce to provide adequate safety oversight for emerging technologies including positive train control (PTC), a communications-based system designed to prevent train accidents caused by human factors…. FRA is a small agency relative to the railroad industry, making the railroads themselves the primary guarantors of railroad safety. Based on our work to date, FRA has about 470 inspectors in its headquarters and regional offices, in addition to about 170 state inspectors. In contrast, the U.S. railroad system consists of about 760 railroads with about 230,000 employees and 200,000 miles of track in operation.”


“Individual Freight Effects, Capacity Utilization and Amtrak Service Quality”
Betty Krier; Chia-Mei Liu; Brian McNamara; Jerrod Sharpe. Transportation Research Part A: Policy and Practice, Vol. 64, June 2014, 163-175. doi: 10.1016/j.tra.2014.03.009

Excerpt: “We hypothesized the existence of a link between individual freight effects and Amtrak’s service quality in the post-deregulation rail industry. We tested the hypothesis with train delay models for long- and short-distance routes. Based on monthly panel data on 1117 directional Amtrak station-pairs for FY 2002 to 2007, we found that individual freight railroads had important effects on Amtrak delays. We also identified other significant delay causes. Among these, the capacity utilization rate, maintenance-related slow orders, and turn points are particularly important because of their potential to be improved upon through stakeholder actions. These findings also have policy implications. Despite the statute giving Amtrak trains priority on freight infrastructure, delays still differ significantly depending on the host railroad. It could be beneficial to look for solutions that further address the potential conflict of interest between freight railroads and Amtrak. For example, there is probably room to increase the effectiveness of the structures for incentive payments by Amtrak to the freights to reduce Amtrak’s train delays.”


“Railroad Safety: Amtrak Is Not Adequately Addressing Rising Drug and Alcohol Use by Employees in Safety-Sensitive Positions”
Office of the Inspector General, Amtrak. Report No. OIG-E-2012-023, September 2012.

Excerpt: “Amtrak’s HOS employees are testing positive for drugs and alcohol more frequently than their peers in the railroad industry. Our analysis of Amtrak’s random drug and alcohol test results shows that these employees have been testing positive for drugs and alcohol at a rate that has been generally trending upward since 2006, and this rate has exceeded the industry average for the past 5 years. The majority of Amtrak’s positive tests since 2006 were for drugs, primarily cocaine and marijuana. In 2011, Amtrak had 17 positive tests for drugs or alcohol, which resulted in a combined positive test rate that was about 51 percent above the industry average, its worst rate since 2007. The 2011 rate was driven by a relatively large number of positive tests by signals and mechanical employees that were both over four times the rate of their peers in the industry. Based on the random test data, we calculated, with 95 percent confidence, that if all 4,454 HOS employees had been tested in 2011, between 21 and 65 of these employees would have tested positive for drug use, with a best estimate of 43 employees. We also calculated that between 4 and 32 of Amtrak’s HOS employees would have tested positive for alcohol use, with a best estimate of 18 employees.

Amtrak is not exercising due diligence to control the use of drugs and alcohol by these employees. Until we presented Amtrak’s key senior management with our preliminary results, they were unaware of the extent of drug and alcohol use by these employees. Further, senior management is not actively engaged in the program, nor have they demonstrated that controlling drugs and alcohol is a clear priority at Amtrak, thereby making it difficult to manage the risk that drug and alcohol use poses to its employees, passengers and the public. Amtrak also did not adequately address, for several years, FRA’s concerns about Amtrak’s program to physically observe HOS employees for signs and symptoms of drug and alcohol use. Consequently, FRA has stated that it may elevate enforcement actions against Amtrak up to and including fining Amtrak in the future if the number of observations is not improved. This may become more challenging because the number of HOS employees requiring observation may increase by 2,260 in 2013 due to potential changes in the regulation.”


About the Authors