Theodore Glickman - Comments

COMMENTS BY THEODORE S. GLICKMAN, Ph.D., PROFESSOR, DEPARTMENT OF DECISION SCIENCES, GEORGE WASHINGTON UNIVERSITY SCHOOL OF BUSINESS ON THE INTERIM FINAL RULE, “HAZARDOUS MATERIALS: ENHANCING RAIL TRANSPORTATION SAFETY AND SECURITY FOR HAZARDOUS MATERIALS SHIPMENTS; RAILROAD SAFETY ENFORCEMENT PROCEDURES; INTERIM FINAL RULE AND PROPOSED RULE” 49 CFR PARTS 172, 174, AND 209 DEPARTMENT OF TRANSPORTATION PIPELINE AND HAZARDOUS MATERIALS SAFETY ADMINISTRATION FEDERAL RAILROAD ADMINISTRATION Docket No. PHMSA-RSPA-2004-18730 RIN 2137-AE02 [Page numbers citations below are to the Federal Register version of the Interim Final Rule from April 16 2008, “Implementing recommendations of the 9/11 Commission Act of 2007” – p. 20755 ff.] http://edocket.access.gpo.gov/2008/E8- 8185.htm I appreciate the opportunity to submit comments on the proposed Interim Final Rule. As a long-time researcher in the analysis of risks related to hazardous materials transportation and the related issue of risk-avoidance routing, I have participated in deliberations about re-routing hazardous materials shipments in various contexts over the years, including catastrophe avoidance, derailment reduction, aquatic ecosystem protection, and, most recently, terrorist attacks on tank cars in potential target areas. Although the initiating events and the consequences may differ from one context to another, the conclusions remain the same: the costs and benefits of rerouting must be considered on a case by case basis; the estimation of costs and benefits must be determined objectively based on all the available evidence; and there will be situations in which the cost-benefit balance tips in favor of re-routing. From the perspective of train accidents, research has shown that re-routing tank cars to avoid urban areas can, in some cases, reduce the risk of hazardous materials releases with only a minor increase (or even a reduction) in the total length of the route. The numbers in the attached table provide some representative estimates of the impact of re-routing in terms of population exposure (a benefit factor) and route length (a cost factor). For each of the six target cities shown, the most likely current route for the selected origin-destination pair is compared to a feasible alternative route that bypasses the target city. These statistics were generated using PC*HazRoute, a commercial software package. From the perspective of homeland security, the imperative to consider re-routing is even more compelling than it is from the train accident perspective. The reason is that an alternative route that avoids attractive targets altogether (including iconic structures such as monuments, buildings, or bridges) would reduce the terrorism risk to zero by removing any incentive for attack. In other words, re-routing can be a mechanism for threat reduction. Of course, there may be trade-offs to be addressed between safety and security when comparing alternative routes, although it would not be surprising to find win-win situations in which the safest route is also the most secure route. The Interim Final Rule provides a framework for analyzing alternative routes that in my view is ill-conceived and, in some ways, overly prescriptive. It does not promote objectivity or rigor in the process of cost-benefit analysis required for evaluating alternative routes. I view it as unworkable and unlikely to result in adequate protection of target cities because it will not promote well-founded re- routing decisions. Some key considerations are the following: 1. DOT’s regulatory proposal to give individual railroads 27 new factors to consider in routing analysis and selection is counterproductive. It is likely to dilute the effort, which instead should be focused on determining a smaller number of highly influential factors (in accordance with the so-called Pareto principle, also known as the 80/20 effect). The number 27 far exceeds any number of factors used in the research literature. The emphasis should be placed instead on impartially identifying the most important factors and developing the database required to evaluate those factors. As it stands, spreading the effort across so many factors is likely to prolong and weaken the analysis unnecessarily. 2. Of even greater concern is the prospect that the routing decisions will be based on subjective weights determined solely by the railroads, with no federal standard or even guidance as to how to do that in a meaningful way. Without an intelligible standard for the decision process, it will not be possible to judge the acceptability of the results. Moreover, the rule does not provide for any independent or systematic review of the routing analysis methodology and selection results. The research literature on multi-criteria decision making stresses the importance of developing weights that truly reflect the preferences of the decision maker and the need to correctly generate and normalize such weights (see, for example, Decision Aids for Selection Problems by D.L. Olson, Springer, 1996). The rule provides no assurance that such best practices will be followed. 3. The notion of delegating routing decisions entirely to the railroads provides a very narrow basis for decision making. There are other stakeholders, including DOT, that ought to be involved in the decision process. The approach advocated here differs starkly from the one taken by the Department of Energy to involve states and tribes in deliberations about the routes for nuclear spent fuel shipments, and by DOT itself in issuing guidance to state officials for the designation of highway routes for hazmat shipments. It also differs from the approach the federal government is using to regulate other modes of transportation given the threat of terrorism, as demonstrated by the Department of Homeland Security’s oversight of the airline industry. I would argue that the level of government involvement ought to be similar here, in order to counter the threat of tank cars being used as weapons of mass destruction. 4. In this rule, DOT has chosen not to require railroads to cooperate in seeking interchange agreements with each other for the sake of risk reduction. If the seven major railroads and the hazmat haulers among the many short line railroads are only going to consider the lines “over which they have authority to operate,” then enormous opportunities will be missed for identifying much lower risk routes—and possibly shorter routes—that would be feasible but for the absence of existing interchange agreements. The implications of this weakness were made clear during the recent events surrounding the re-rerouting of tank cars of chlorine around Washington, D.C. The railroad involved declined to participate in an interchange agreement with another railroad that would have made possible a sensible re-routing on a non- target route 50 miles to the west. Instead, it agreed only to refrain from using its north-south line while continuing to use its east-west line, which still comes close to the U.S. Capitol. If barred from using that line, the railroad maintains that it will have to resort instead to using an implausible route that would incur hundreds of extra miles and pass through Cincinnati, Cleveland, Buffalo, and Albany, and the metropolitan area of New York. I urge DOT to reconsider the hands-off approach reflected in this rule and to require the railroads to undertake a meaningful initiative to consider the possibility that re-routing, when approached with an open mind and subjected to scientific scrutiny, can in certain circumstances be an effective form of counterterrorism. Otherwise, we run the risk that national security expert Stephen Flynn of the Council on Foreign Relations will be correct in saying that: “When the next big attack occurs on American soil, the public will be appalled to learn how little has been done to prevent it.”