Holistic

Modern transportation systems have pervasive and far-reaching effects on society and the environment. Mobility and other benefits of modern transportation arrive with many, serious undesired consequences: deaths and injuries in transport accidents, pollution of air, water and groundwater, noise congestion, greenhouse gas emissions, to mention but a few. In this book leading international researchers explore the issues and concepts and define the state of knowledge concerning transportation’s full costs and benefits.

Individual works of excellence

As mentioned, 20 or so academics, from around the world, including the USA, Germany, Australia and France, have submitted their own ideas on costs and benefits. Displaying my ignorance, I have to admit I don’t know that many of them; perhaps I should, but I recognize their respective universities as leading institutions in the field of transportation. Two of them I do know: I have met Douglass Lee of USDOT in person, when attending the  TRB Annual Meeting in 2004. He is head of the Transportation Economics committee at TRB. And I’ve run into David Hensher, ETC 2005 I think it was. David Hensher is one of the major  figures in transportation research, so if both Douglass and Hensher are contributors than this IS a good book. Well, maybe they were not so famous 12 years ago, but ceratinly now.

Outdated?

The book is not new. It was first published in 1997, and although the empirical data may no longer be anywhere near true, the principles and methodologies still hold.

amazon.com

• Buy this book at amazon.com

Related

• husdal.com: Book review: Transport – an economics and management perspective

Originally written in April 2004, revised and updated in March 2009

1 Introduction and outline

Independent thought, critical analysis, individual judgement and expertise in the application of research theory and practice are but a few of the ingredients necessary to formulate, develop, carry out and successfully complete any kind of research work. Unfortunately, many researchers are unaware, or worse: blatantly ignorant, of the underlying tacit assumptions that operate in their field of research and under which they work. This paper will address this issue for the author’s own research on reliability and vulnerability as input parameters in cost-benefit analyses. Thus, the aim of this paper is to clarify which scientific approach or which school of thought this author is inclined to follow in his research, and how this has affected and will continue to affect his line of research.

A research subject that sees its aim as decision support, balancing different sides and arguments, rather than holding one objective truth, has much in common with the explorative nature of Lakatosian research programmes. At the same time it has no overall common assumptions or paradigms it needs to adhere to in Kuhnian sense: Reliability analysis is deductive in nature; vulnerability adds a Bayesian vagueness to reliability. Costs and benefits are firmly set in the utilitarian thought of classic economics, implying society as the sum of rational individuals. Combining these diverging views is challenging, but not insurmountable, if following Giordano Bruno’s ideas of pictorial concepts, rather than developing intricate formulae.

Firstly, the scientific and historical backdrop for the terms reliability, vulnerability and cost-benefit analysis as they are used in the henceforth research will be described. Secondly, different approaches to knowledge in resolving the methodological and epistemological problems within the research will be evaluated. Thirdly, and finally, the author’s own approach and insights will be stated.

2 Defining the object of analysis

The working title of the research project is “when you cannot get from here to there” or in proper academic terms “reliability and vulnerability versus costs and benefits”. It aims at investigating the most typical factors that affect the reliability, and particularly, the vulnerability of the transportation infrastructure and in consequence establishing a framework for how these factors should be approached in terms of costs and benefits.

Rationale

The rationale for the research is that the reliability or vulnerability of the transportation network is only seldom looked at as a decision parameter in cost-benefit analyses, especially for new projects. For the most part, saved travel time is what drives the policy of the decision makers, increased reliability is as a rule not a subject for closer evaluation; it is simply taken for granted. By adding reliability and vulnerability to the traditional equations of costs and benefits it is hoped that transportation planners and professionals will not only consider economical arguments, but also dare to take on political statements that may be in opposition to strictly factual costs and benefits. In essence, three fields or subjects are brought together, engineering (reliability and vulnerability), economics (cost and benefits) and politics (decision making). The purpose of this essay is to bring forth and highlight the epistemology, ontology and methodology of the three fields that make up the backdrop for this research, and to look at how their tacit assumptions will affect the undertaking of the research that is to follow.

Is this research scientific?

Webster’s Dictionary defines “science” as “the state of knowing: knowledge as distinguished from ignorance or misunderstanding”, and everyday connotations often associate science with “objective knowledge” or “absolute truths” about an object of study, a knowledge that not necessarily needs to be acquired through personal experience. Knowledge applies to facts or ideas acquired by study, investigation, observation, or experience. With the author’s engineering, and thus commonly established “scientific” background in mind, it is only natural for him to search for some objective knowledge in his chosen field of research. The Oxford Advanced Learner’s Dictionary defines science as “knowledge about the structure and behaviour of the natural and physical world, based on facts that you can prove”. The question then arises to the author: Firstly, does there or does there not exist a factual and provable knowledge about his field of study, why, or why not? Secondly, where in the realms of the sciences or knowledges is this research situated and from which perspective is he (am I) approaching his (my) research? These are the question that this essay will attempt to answer.

Where did it all start? The author’s initial interest in his particular research was stirred by the research work done by Katja Berdica for her PhD (Berdica, 2002) at the Royal Institute of Technology in Stockholm, Sweden. Her thesis becomes much of the bedrock foundation for the undersigned author, as he picks up where Berdica left:

Road vulnerability analysis could thereby be regarded as a hub for the whole battery of transport studies needed to gain the insights necessary to describe how well our transport systems work in different respects, what steps to take and what policies to implement in order to reach desired goals. (Berdica, 2002, p.127)

It is tempting to then view vulnerability analysis in terms of the Lakatosian research programme, spinning off into various strands (batteries) of research, each strand developing different or diverging insights, from which only the fittest will survive. At the same time, there is also a concept of converging insights, and Berdica suggests that:

[...] vulnerability in the road transportation system should be brought out and recognized as a crucial part in the infrastructure, and [...] be the meeting point for all the different strands of transport reliability research and other issues. (Berdica, 2002, p.127)

Such a view acknowledges that an interdisciplinary approach is necessary to bring the research forward and to develop new ideas. Subjecting hard core (transportation network) reliability to interdisciplinary critique will undoubtedly allow for the protective belt to be tested in order to generate new ideas. It is hoped that the research that underlies this essay will provide the necessary drive to turn transportation network reliability studies into living and generating, not degenerating, research.

Transportation network analysis has long has its roots in the natural sciences and civil engineering departments, and traditionally, when looking at the reliability of transportation networks, it is done with a systems engineering approach. Reliability is here dependent upon functioning or non-functioning links within the network, and may be defined as the degree of stability of the quality of service that a transportation system offers.

More often than not, transportation network reliability is limited to two aspects only, videlicet, connectivity and travel time reliability. Although it conceptually makes perfect sense to sum up reliability in these two variables, there are an almost unlimited number of factors that may or may not contribute to the final value, and thus, should be part of the overall equation. Thus, an interdisciplinary Lakatosian research programme approach is definitely required.

From a user perspective, what matters most in relation to a transportation network is its availability at the given time of travel, in other words, can I get from A to B by using the intended route and means of transport, at the desired time of travel, which would be the best case. Or, there exists no route or means of travel at all, which can take me from A to B at the desired time of travel, which to the user would be the worst case. Consequently, the probability of the network of being available, accessible and expressing the desired level of service may thus be used as a measure of reliability. Conversely, the road network’s susceptibility to not being available, accessible and expressing the desired level of service may thus be construed as a measure of vulnerability.

3 Reliability, vulnerability, costs, benefits and decisions

Reliability

Robert Lusser is by many regarded as the father of reliability, and Lusser’s law is perhaps the most quoted and used equation in reliability engineering. It defines the reliability of a series system as the combinatorial product of the individual component reliabilities. Formulated during WWII, it established the notion that no system is stronger than its weakest component, and was first used to improve the success rate of the German V-1 bombs that rained over England.

The reliability of any object as such is most commonly described as the probability that this object will function (or fail) according to an object-specific set of standards. Put very crudely, it is the ratio of observed functioning and observed non-functioning events. If an item functions in 95% of the observed cases, then it is 95% reliable, or in other words, the probability that the item will function is 0.95, since reliability by convention is described as a number between 0 and 1 rather than in percentage numbers. The literature on reliability is immense, as it draws upon statistics and probability theory, both immense subjects of their own. Suffice it to say that probability, and thus reliability, take their roots from deductive reasoning based on statistics. Probability theory traditionally presupposes classical set theory/classical logic, relying heavily on deductive reasoning. Does reliability (probability) theory qualify as science? Yes, in the most classical sense it appears to be, because it claims to absolutely determine the probability of the functioning of an object. If an object is 95% reliable, it will function 95 out of 100 times. Consequently, in maintenance practice, the object is replaced before it reaches its lifetime of 95 probable ons and offs. Nevertheless, the object could fail already long before its probable lifetime, or go on working well beyond it. In this sense, probability turns into possibility, and no longer is a science.

Statisticians mostly base their assumptions on observations of real world events, from which certain predictions (future events) or correlations (past events) are made. If the deductions are correct, the theory will live up to the experience, or vice versa. However, how can one be certain that what is observed represents the full truth about the world; or if not the world, then at least the truth about the subject of interest? This is where Bayes’ theorem comes to play an important role. Bayes’ theorem is used in statistical inference to update estimates of the probability that different hypotheses are true, based on observations and knowledge of how likely those observations are, given each hypothesis. In essence, in applying a Bayesian approach to probability, and thus reliability, it is possible to come up with a degree to which we feel our predictions are true (probable). Bayesianism is the philosophical tenet that the rules of mathematical probability apply not only when probabilities are relative frequencies assigned to random events, but also when they are degrees of belief assigned to uncertain propositions. Bayes’ theorem thus becomes a way of fitting reliability with a less absolute notion, while still upholding the scientific credibility of delivering the one true answer.

Vulnerability

Berdica defines the vulnerability of a road transportation system as the susceptibility to incidents that can result in a considerable reduction in road network serviceability, with reliability describing adequate serviceability under the operating conditions encountered at a given time.

With vulnerability as a complement to reliability, the issue of vulnerability can now be described in the well-established and solid framework of reliability theory. Consequently, the methodology used in analysing vulnerability will be analogous to the methodology used in analysing reliability. This being the case, we may see vulnerability as non-reliability and vice versa.

Again transferring this to the Lakatosian view of research programmes, vulnerability analyses then represent the positive heuristics that that are legitimate and worthwhile pathways of research (Smith, 1998), with the underlying theories of reliability being the hard core holding the key foundational assumptions. Properly and adequately describing what constitutes vulnerability then (in a strange twist of literal irony) becomes the protective belt of auxiliary hypotheses which can be tested and falsified.

Note here that there is one important addition to the traditional reliability approach that appears in Berdica’s application of reliability to describe the vulnerability of a transportation network: Reliability is linked explicitly to the ability of supplying adequate serviceability, not simply to the probability of failure, as in the classic case of reliability studies. The sheer notion of susceptibility implies something more than probable and statistically predictable failure, namely the influence of external circumstances, circumstances that cannot be predicted, and which may, but not necessarily will have an influence on the reliability numbers. In this sense, pure reliability appears as being strongly independent from external circumstances; vulnerability on the other hand is not. Reliability as such is an intrinsic and stable value, vulnerability is variable. This perspective on vulnerability leads to the conclusion that although there can be a quasi-objective measure of reliability, there can not be an objective measure of vulnerability. Consequently, vulnerability is less measurable than, or not even as measurable as reliability, and is thus open to discussion.

This brings us back to the previous statement of research programmes, and it is the aim of this research that bringing vulnerability and reliability into the field of transportation network analysis will generate new debate, new ideas and new research avenues.

Costs and benefits

Jeremy Bentham (1748 -1832) is often credited with being the father of cost-benefit analysis, one of the many tools of both macro- and microecononomical analysis. Bentham is one of the first utalitarians, and devised a means of calculating the consequences of various actions in terms of utility. The underlying principle of his felicific calculus is still applied in modern-day cost-benefit analysis, but has also found its way into politics and economics. Using Benthams felicific calculus it is possible to ascertain which action will cause the greatest good for the greatest number. The felicific calculus was sketched by Bentham in chapter 4 of his Introduction to the Principles of Morals and Legislation, see Burns et al. (1970):

When determining what action is right in a given situation, we should consider the pleasures and pains resulting from it, in respect of their intensity, duration, certainty, propinquity, fecundity (the chance that a pleasure is followed by other ones, a pain by further pains), purity (the chance that pleasure is followed by pains and vice versa), and extent (the number of persons affected). We should next consider the alternative courses of action: ideally, this method will determine which act has the best tendency, and therefore is right.

Since classical utilitarian theory considers the rightness of an action as a function of the goodness of its consequences, the felicific calculus could, in principle at least, establish the moral status of any considered act. In strictly utilitarian terms the felicific calculus allows for society to justify inequality, if this, in sum, generates more happiness than equality. Critics would claim that the felicific calculus fails as a moral theory because there is no principle of fairness. Carried to the extreme, it would be moral to torture one person if this would produce an amount of happiness in other people outweighing the unhappiness of the tortured individual. Critics also point out that the happiness of different people is incommensurable, and thus a felicific calculus is impossible in practice.

Nevertheless, with a lesser moral undertone the basic principles of this method are still used today, and are the linchpin of today’s cost-benefit analysis: If the overall benefits (to society) of any action or decision outweigh the overall costs (to society), then the action or decision is justified, or at least economically sound. On a personal side note it may be worth pondering the welfare system of today’s modern society, which is built on achieving equality for everyone. Consider, for a moment, whether more equality among people also means more happiness among them, or whether inequality in fact is justified, because the happiness of the better-offs outweighs the misery of the worse-offs.

Utilitarianism was revised and expanded by Bentham’s more famous disciple, John Stuart Mill. In Mill’s hands, “Benthamism” became a major element in the liberal conception of state policy objectives. Utilitarianism has been deserted as a moral philosophy in modern day society, but continues to live on in two key areas, in the reasoning of everyday life and in neoclassical economics. Classical economic theory is founded on Bentham’s ideas, where the hedonistic nature of the individual as a pleasure-seeking and pain-avoiding individual (aka consumer) remains unfettered and unchanged by religion, philosophy, moral or culture. More euphemistically sounding but not less hedonistical in nature is Adam Smith’s economical man. At the heart of economic theory is homo economicus, the economist’s model of human behaviour. In traditional classical economics and in neo-classical economics it was assumed that people acted (only and always) in their own self-interest. Adam smith argued that society was made better off by everybody pursuing their selfish interests through the workings of the invisible hand, meaning the ability of the free market to allocate factors of production, goods and services to their most valuable use, everybody acting from self-interest, spurred on by the profit motive

Does cost-benefit analysis qualify as science?

Halvorsen (1989) answers this question in part, initially accepting, based on Chalmers (1982), and then finally discarding, Lakatos’ idea of research programmes as a valid view of how economic reason develops new ideas and theories. The point he makes is that human beings seldom are acting rational and in pure self-interest, and thus rarely are in compliance with economic models. Humans, he argues, act upon the norms and morals of society, often in conflict with any sense of utility-seeking rational behaviour. In a Lakatosian sense, these contradictions in the protective belt are the motive power towards change and development of economic theories. However, these contradictions are usually not testable for every individual, and economists therefore must rely on theory rather than experience.

The most concise and probably most used definition of economics is the study of how society uses its scarce resources. Thomas Carlyle, a 19th-century Scottish writer, coined the phrase “the dismal science” when referring to economics. The phrase “dismal science” refers to the unreliability of economics in comparison to conventional sciences such as mathematics, physics, or biology, the unreliability being the inability to ascertain any clearly-defined laws because in the market some unidentifiable factor always may be influencing a particular event or phenomena. It’s nearly impossible to isolate any given variable in economics, so the dismal science is mainly based on theories. These theories may contradict each other, like efficient market theory and behavioural finance, but they may be proven true in certain cases or even at the same time. Furthermore, when studying economics, evidence often turns out to be coincidence more than a fact. Andreassen (1989) elaborates on this in describing the rational man as a fictional and virtual construct that only exists in the mind of economists, a Frankenstein consisting of bits and pieces of various separate models of behaviour. The economic man then becomes the representation and simplification of a reality that is too complex to be understood and fully explained by economists. If it is the objective of science to deliver knowledge and truth about the (real) world, then economy fails miserably as science. Nonetheless, economists often put a high degree of confidence in their models and predictions, with contradictory evidence as a result. The recent debate over interest rates and deflation in Norway, though unrelated to cost-benefit analysis, may serve as a small illustration here.

A discussion of the rational behaviour of individuals is particularly necessary in the analysis of the reliability of transportation networks. From a systems theory point of view the matter of functioning or non-functioning links seems easy enough. What is often forgotten is that the transportation system is populated by individuals whose behaviour is not easily predicted. Albeit the increasing numbers of road rage suggest that many travellers or drivers act out of self-interest, we can also compare traveller behaviour to the well-known prisoner’s dilemma (Smith, 1998), in adjusting one’s own travel habits by second-guessing other travellers behaviour.

Decision-making

The reason for bringing reliability and particularly vulnerability into the realm of cost-benefit analysis is to have decision makers consider other points of view than purely monetary or “scientifically objective” numerical (i.e. “true”) arguments.

Decision making is a process that leads to a choice between a set of alternatives, and can be split into the following sequential process :

• Defining the decision problem (objective)
• Determining the set of evaluation criteria to be used
• Weighting the criteria
• Generating alternatives
• Applying decision rules
• Recommending the best solution to the problem

Any decision-making process begins with the definition of the problem or the objective to be reached. Once the decision problem is defined, what follows is setting up a set of criteria that reflect all concerns of the problem and finding measures as to which the degree is achieved. The purpose of weights is to express the importance or preference of each criterion relative to other criteria. Alternatives are often determined by constraints, which limit the decision space of feasible alternatives. Decision rules integrate criteria, weights and preferences to generate an overall assessment of the alternatives. Recommendations are then based on a ranking of the alternatives.

However, all decision making has a degree of uncertainty, ranging from a predictable (deterministic) situation to an uncertain situation. The latter one can be subdivided into stochastic decisions (which can be modelled by probability theory and statistics) and fuzzy decisions (which can modelled by fuzzy set theory and others). In some situations, decision making involves the risk or some uncertainty of making a “wrong” decision, because the information acquired is insufficient or the approach used is inappropriate. Using Bayes’ theoretical framework, the uncertainty that is involved may in some cases be quantified and as such add another decision criteria to the evaluation process.

In the end, it is the decision maker, who determines the criteria, the factors, the constraints, the individual weighting and the decision rules. Decisions based on reliability and probability have a scientific aura around them, being absolute by number; decisions based on vulnerability are more relative and descriptive, not absolute, since vulnerability is associated with susceptibility, meaning the state of being very likely to be influenced, harmed or affected by something. This is an individual issue, not easily translated into collective terms. Then again, the aim of this research is not to find a justifiable measure of vulnerability, but to point at vulnerability as a parameter that needs to be included in cost-benefit analyses and to show how this can be done. This being the case, this research warrants only scientific connotations, not scientific denotations.

4 Implications for the research and development of new ideas

Bentham’s felicific calculus was said to be impossible in practice, because the (different) happiness of different people was incommensurable and could not added together into a single value. Yet this calculus is exactly what is applied in today’s cost-benefit analyses, and is rooted in the economic theory of supply and demand. The demand of each individual is added together as society’s total demand for a good at a given price. In similar manner, the demand representing the value that each individual attributes to this good, representing each individual’s valuation of the good is added together as society’s total valuation of the good. This is a representation, not a fact. Consequently, it must be kept in mind that cost-benefit analyses are a simplification of complex sets of value and appreciation, not hard core facts.

Mc Closkey’s now classical article on The rhetoric of economics (Mc Closkey, 1983) contends that economics, in essence, like the proverbial emperor, has positively no clothes on, yet economics has always claimed and continues to claim that it is steeped in science. Mc Closkey goes as far as suggesting that statistical significance, regardless of how useful it is as input into economic significance is not the same thing as economic significance. Hajek and Hall (2002) note that hypothesis testing at a constant significance level is inconsistent with Bayesian inference and decision theory, and thus both classical and Bayesian statistics need to be properly integrated into a more general theory of decision. This is also noted by Aven (2003) in Foundations of Risk Analysis. Making good decisions is to a large extent about the quality of the decision-making process, the definition of the problem, the overall goals, criteria and preferences, the identification of alternatives, the use of risk and decision analyses, and the managerial review and judgement. There are no objective ways of measuring this quality, but one can of course be more or less confident about the process underlying the decision. Taking proper considerations towards all parties is not always easily translated into mathematical formulae, and perhaps it shouldn’t be. In dealing with societal considerations more can probably be gained by deliberations, for people to confer, ponder, exchange views, consider evidence, reflect on matters, negotiate and attempt to persuade each other. This is the hard core of this author’s research.

In the concluding paragraphs in their article on induction and probability, Hajek and Hall (2002), suggest the need for more cross-fertilisation between Bayesian and classical statistics, as seen in Mayo’s theory of error statistics The centrepiece of Mayo’s argument is the notion of severity, referring to the notion that any method or procedure of testing also must consider how the data were generated, modelled, and analysed to obtain relevant evidence in the first place.

Hajek and Hall (2002) foresee that

in this brave new world of inter-disciplinarity and rapid communication, inferential methods developed within one filed are increasingly likely to be embraced by practitioners of another.

Devising a methodology that allows for the inference of reliability and vulnerability is one of the tasks in this research. The inferred values of reliability and vulnerability are then used in cost-benefit analyses.

The purpose of a cost/benefit-analysis is to weigh the costs of a proposed project against the benefits of the project. If the benefits exceed the costs, then the project increases society’s welfare. If the costs exceed the benefits, society will experience a loss of welfare. The argument of vulnerability versus reliability is analogous: Any vulnerability of the transportation system causes disruptions that cause costs, which are in fact a loss of welfare; vulnerability is thus a cost that is quantifiable. If society puts measures in place to reduce the vulnerability of the transportation network, the increased reliability represents a benefit.

One question that remains open at this stage in the research is whether the inferred values or measures of reliability and vulnerability are in fact commensurable or not. Then again, keeping the above argument of confidence of arguments at hand, it may not be required, if decision support is an analytical tool providing well-founded information and not final answers.

5 Conclusion

The initially stated purpose of this essay was to bring forth and highlight the epistemology, ontology and methodology of the three fields that make up the backdrop for this research, and to look at how their tacit assumptions will affect the undertaking of the research that is to follow.

Whether or not this has been critically evaluated to the point remains to be seen in the forthcoming research. Nonetheless, the essay has raised the awareness of not entering headfirst into a research without first acknowledging and appreciating the historical and philosophical timeline the topic has evolved under.

The aim of this paper was to clarify which scientific approach or which school of thought this author is inclined to follow in his research. Three fields or subjects are brought together, engineering (reliability and vulnerability), economics (cost and benefits) and politics (decision making). The idea behind the research is to blend statistical, economical and political arguments in order to achieve a novel and unifying framework for decision making within transportation planning.

In summing up, this research may very well be a discourse of its own, not fitting into any particular of the traditional contexts that surround the three fields or subjects. What it tries to do is to blend statistical, economical and political arguments in order to achieve a novel and unifying framework for transportation planning. Reliability and vulnerability are seen as crucial input parameters underpinning all further research. The probability of the network of being available, accessible and expressing the desired level of service may be used as a measure of reliability, and conversely, the road network’s susceptibility to not being available, accessible and not expressing the desired level of service may thus be construed as a measure of vulnerability.

Rooted in reliability theory, objective truths or numbers regarding the reliability of the road network are used to assess its possible, yet not necessarily probable, vulnerability, because susceptibility by definition is not always predictable. In this research, reliability or probability is subjected to its Bayesian susceptibility to uncertainty, and a vulnerability analysis takes advantage of and exploits this uncertainty. Bearing in mind that we cannot be certain of the absolute value of the reliability of a road network, the utility to society regarding reliability must also be viewed as uncertain, and so must the ratio of costs versus benefits. The translation of qualitative measures into quantitative measures that is necessary to perform a cost-benefit analysis already involves an element of uncertainty in itself, “worsening” and weakening its objectivity, if the numbers for cost and benefits are addressing vulnerability or reliability. What this goes to show is that there can be no objective measures regarding vulnerability, and thus no objective decisions, in such that any decision to mitigate vulnerability is a qualified guess rather than a well thought-out process. That said, there’s nothing wrong with qualified guesses.

Vulnerability analyses present predictions and uncertainty assessments of observable quantities and provide decision support, not hard core decisions. The analyses need to be put into a wider decision-making context, which we may refer to as a managerial review and judgement process, and this process then results in a decision.

A research subject that sees its aim as decision support, balancing different sides and arguments, rather than holding one objective truth, has much in common with the explorative nature of Lakatosian research programmes. At the same time, it has no common assumptions or paradigms it needs to adhere to in a Kuhnian sense.

Nearing the final conclusion of this essay, in attempting to unify statistics, economics and politics, probably (not meant in a scientific statistical sense) the best guideline for the henceforth research is to follow Feyerabend’s golden rule of “anything goes”. Without a fixed ideology that can comprehensively span 3 fields of research the only approach which does not inhibit progress and discovery of new knowledge is in fact “anything goes”.

anything goes’ is not a ‘principle’ I hold [...] but the terrified exclamation of a rationalist who takes a closer look at history.

If rationalism is considered being a philosophical doctrine that asserts that the truth should be determined by reason and factual analysis, then “anything goes” implies that whatever reason constructs and manages to justify as factually true, is what in the end establishes itself as the truth, until a new or different truth is established and justified. In the prospects of this research such a statement means admitting that one’s own research sooner or later will be inadequate and insufficient as grounds for the kind of decision support it is supposed to deliver.

In finishing this paper and preparing for the vexing critique of my contemporaries the words of Giordano Bruno came to cross my mind, as he was facing the his trial in 1600, before being burned at the stake for heresy:

Maori forsan cum timore sententiam in me fertis quam ego accipiam. Perhaps you, my judges, pronounce this sentence against me with greater fear than I receive it. Giordano Bruno ( in: White, 2002)

The philosophical kinship that this author feels toward Giordano Bruno is indeed a fitting finale to this research essay. Giordano Bruno was a philosopher who took the current ideas of his time and extrapolated them to new and original vistas. He claimed that all matter was intimately linked to all other matter, that we live in a universe in which all things are related. Bruno is generally not considered a scientist, yet, ironically enough it is perhaps the scientific element of Bruno’s work that presents us with the most lateral links between his ideas and modern thinking. Bruno was never in any sense a practical researcher, to quote White (2002); he did not think in terms of mathematics or experiment. In fact, he actively disapproved of the way the new science of his time was becoming increasingly entwined with mathematical proof and purity, much in the same manner that this author disapproves of the way cost-benefit analysis represents utility as an economical and mathematical measure that can be compared against other economical and mathematical measures.

Giordano Bruno is unique compared to the other martyrs of his time because of the power of his forward-thinking, where others were personal-thinking or contrary-thinking. Bruno held a broader vision; his heresy was all-embracing. He defended the rights of all humans to think as they wished; he offered an alternative to the ideas enforced by orthodoxy. He was a man who wished to steer humanity towards reason, who wanted to allow us to conceptualise freely rather than have our thoughts determined for us, free from Kuhnian paradigms, free from the hard core assumptions of Lakatosian research programmes.

Bruno advocated the use of conceptualising, that is to think in terms of images. He said that to think was to speculate with images. Complex scientific correlations are often better explained in pictures than in mathematical formulae. Consequently, Bruno was able to rationalise his theories, even though he used no mathematics. Bruno’s vision of picture logic is used by almost anyone in the industrialised world today; just take Windows ® software as an example. The workings of Windows are (as is obviously seen by most users) based upon logically connected images.

Interestingly enough, Bruno predated Karl Popper by three and a half centuries when he wrote:

He who desires to philosophise must first of all doubt all things. He must not assume a position in a debate before he has listened to the various opinions and considered and compared the reasons for and against. He must never judge or take up a position on the evidence of what he has heard, on the opinion of the majority, the age, the merits, or prestige of the speaker concerned, but he must proceed according to the persuasion of an organic doctrine which adheres to all real things, and to a truth that can be understood by the light of reason. (Guiordano Bruno, De immense, De monade, De minimo, 1591)

Rather than spinning his ideas from the yarn of algebra, the cobweb of modern science, Bruno moulded pictures and manipulated visual images to interpret complex ideas. This, by coincidence, comes close to McCloskey’s metaphor-laden rhetoric of economics, and makes the argument that economics presents itself in a Brunoan manner, while attempting to uphold its scientific bedrock of mathematics.

The research that this author is carrying out aims at taking reliability and vulnerability into the realm of cost-benefit analysis to serve as decision support. Decisions should not be determined by numbers alone; decisions should be fully envisioned and comprehended by the decision makers. This is only possible by speculating with images what the outcome of the decision will be. Following Bruno’s lead, leaving the mathematical world of realibility (probability) and cost-benefit (economics) behind, the research that is to follow intends in every way to be rich in images, but poor in formulae.

6 Bibliography

Andreassen, L. (1989) ‘The representative decision maker: a Frankenstein encounting reality’, in: K.A. Brekke and A.Torvanger (ed.) Philosophy of Science and Economic Theory, Central Bureau of Statistics in Norway

Aven, T. (2003) Foundations of Risk Analysis : A Knowledge and Decision-Oriented Perspective. John Wiley & Sons.

Berdica, K. (2002b) An introduction to road vulnerability: what has been done, is done and should be done. Transport Policy, 9 (2), 117-127

Burns, J.H. and Hart, H.L.A. (1970) The collected works of Jeremy Bentham: An introduction to the principles of morals and legislation. The Athlone Press, University of London.

Chalmers, A.F. (1982) What is this thing called Science? Open University Press.

Hajek, A. and Nall, N. (2002) ‘Induction and Probability’, in: P. Machamer and M. Silberstein (ed.) The Blackwell Guide to the Philosophy of Science, Blackwell Publishers, pp. 149-172.

Halvorsen, K. (1989) ‘Lakatos as a motive power in economics’, in: K.A. Brekke and A.Torvanger (ed.) Philosophy of Science and Economic Theory, Central Bureau of Statistics in Norway

McCloskey, D. (1983) The Rhetoric of Economics. Journal of Economic Literature, 21, June: pp. 481-517

Smith, M.J. (1998) Social Science in question. Sage Publications, UK. • Walliman, N. (2001) Your research project: a step-by-step guide for the first-time researcher. Sage Publications

White M. (2002) The Pope and the Heretic. HarperCollins Publishers

Related

• husdal.com: Broader research = better research?

The big question

Admittedly, the topography of Norway makes building roads, or any transportation infrastructure for that matter, quite a challange, but if other countries can pull it off, why not Norway? The answer, it seems, is a political one. It is not the planning authorities or the central government who decides, but the local politicians. To put it simple, what in the US is known as “pork barrel spending” is what rules many of Norway’s infrastructure development projects. Why?

Votes Count but the Number of Seats Decides

The first major discussion of this topic came in 2006, in Knut Boge’s PhD dissertation at The Norwegian School of Management:

Why has Norwegian authorities pursued a road policy contrary to most other West European industrialized countries? Why were highly noticeable congestion, accident and environmental problems within and near Norway’s major population clusters overlooked or ignored for decades?

The answer he says, lies in the way Norway decided about infrastructure investment, where prevalent legislator rule paved the way for a road policy governed by political rather than economic and technocratic logics:

Oil revenues made the Norwegian State nouveau riche from the early 1980s. But Stortinget (the Norwegian Parliament) shifted partly the State’s responsibility for financing trunk roads, motorways and other highways to the counties and private actors from 1985 through a pork barrel deal that imposed common turnpike financing. The 1963 Road Act that explicitly stated financing and construction of trunk roads, motorways and other highways as State responsibilities was not amended. Road financing through local turnpike companies instead of or in addition to State road appropriations made those who profited from turnpike projects important road political players.

What this implies is that local politicians have a much greater influence on infrastructure projects than in many other countries. Often there will be competing projects in nearby locations, and no oversight decision.

Geographical redistribution with disproportional representation

Knut Boge’s work was later followed up by professors Leif Helland and Rune Sørensen in their article Geographical redistribution with disproportional representation, stating that swing voters play an important role:

To some extent, individual districts are able to get their desired projects approved by parliament. But there is more to road investments than simple district demand. National parties allocate more road expenditures to districts with many voters and coordinate districts’ demands to win seats in parliamentary elections.

To sum it up, roads are not built where most people are, but where most voters are.

Conclusion

What is the result? The politically attractive projects win over the economically attractive (or sound projects).. From a cost-benefit perspective, mot of the time this is the wrong decision. And as long as our electoral system is the way it is, I don’t think it will change much. Unfortunately.

Another take on this story

(Update 2009/10/02) I recently discovered another blog posting on the exact same topic, and polemarchus.net provides a much better and probably more eloquent discussion Norwegian roads and swing votes than what I have given above.

The most disturbing point isn’t that the distribution of money is non-optimal from a cost-benefit perspective. That’s the nature of politics. The big problem is that there is a skewed distribution as a result of election strategy concerns. [...] Valuing decentralized communities highly is acceptable from a democratic point of view. Consistently bribing swing voters with public money isn’t.

References

Helland, L., & Sørensen, R. (2008). Geographical redistribution with disproportional representation: a politico-economic model of Norwegian road projects Public Choice, 139 (1-2), 5-19 DOI: 10.1007/s11127-008-9373-z

Links

• tu.no: Veikroner går til å vinne velgere
• aftenposten.no: Veikroner går til å vinne velgere
• aftenposten.no: Kraftige kutt i veibevilgninger
• polemarchus.net: Norwegian Roads and Swing Voters

Related

• husdal.com: Articles tagged with samferdsel
• husdal.com: Norwegian roads are so slooooow

Politics?

Part of my dayjob is to do Quality Assurance of major investment projects. Established by the Norwegian Ministry of Finance in 2000, the Quality Assurance Scheme is performed by external consultants under a framework contract, providing a second opinion on major investment projects. So far so good. What is not so good is that our second opinion isn’t always heard, as is the case of rail versus road in Bergen, where the question was whether to build a 2nd parallel rail tunnel because the existing tunnel was stretched beyond capacity or whether to build a road tunnel  because the road network is utterly congested. From an overall cost-benefit perspective, building a  road tunnel was preferable to building a rail tunnel, yet the government decided to go for  rail, and not in favor of the road. Why? Personally I think it is because politically, rail is environment-friendly and inline with the climate change mantra of reducing car traffic and increasing ridership on mass transit. Nothing wrong with that, but what about when it is not economically sound to do? Or maybe it is the economists who have it wrong?

Why QA?

I should add that I was not involved in the particular QA of said project. It just came to my mind when I read this article (in Norwegian), where the head of Jernbaneverket, the Norwegian National Rail Administration, expresses his satisfaction at the governments decision to listen to his particular needs rather than listening to the economists who disqualified the rail project and decided in favor of the road tunnel.

In the end…

…public investment projects are always political decisions, are they not? Is QA even necessary?

Trading operation costs against disruptions costs

In Reliability Models for Facility Location: The Expected Failure Cost Case, L.V. Snyder and M. S. Daskin do just that. They present a model that calculates the minimum location cost for multiple facilities, while at the same time minimizing disruption costs. The model generates a trade-off curve between day-to-day operation costs and expected failure costs, showing that substantial reduction in expected failure costs can be achieved with minimal increasing in operation costs.

Methodology

For the more analytically inclined I will just say that the model is based on the classic P-median problem (PMP) and uncapacitated charge location problem UFLP and a Langrangian relaxation algorithm. The model considers the failure costs, given a probablity that each facility will fail and assuming that multiple facilities can fail simultaneously. The model by Snyder and Daskin shows that even a small increase in operation costs and a careful selection of facilities can lead to significant reductions in failure costs.

Practical background

The background for developing the model is that, obviously, very few firms would be willing to choose locating its facilities such that the normal operating costs are higher than optimal, just to hedge against the case of potential disruptions, especially if you don’t know where the disruption will occur or what kind of disruption that will occur.

Theoretical background

The model synthesizes three strands of research: Network reliability, covering problems or models, and supply chain vulnerability and resilience.

More suppliers is not more reliability

Interestingly, what should be remembered in selecting the optimal number of facilities vis-a-vis potential supply chain disruptions, is that adding redundant suppliers does not create more reliability, because it in fact increases the overall probability that one or more suppliers will fail.

The P-median problem (PMP)

The P-median problem minimizes a weighted sum of the operating cost and the expected disruption costs, given a random distribution of facility failures. Certain facilities can be designated as nonfailable and multiple failures can occur at the same time.

The reliability fixed-charge problem (RFLP)

The RFLP forms the basis for the UFLP, in that the RFLP can only minimize costs for a different set of locations, whereas the UFLP can accommodate unlimited facilities, and thus allows for the construction of additional facilities where this is most favorable for the overall solution.

Results

The results show the tradeoff between UFLP costs (operation costs) and expected failure cost, e.g. at an operating cost of $1,000,000 one can expect a failure cost of $400,000. Increasing operating cost to 1,200,000 reduces failure costs to $200,000.  The model can also show how many nonfailable facilities that are needed for a certain level of reliability.

Conclusion

The paper makes a convincing argument that facility location can be of critical importance in in reducing the cost of disruptions. The weakness though, as I see it, is that the model assumes added transportation cost as the driving factor, and albeit this probably IS the driving factor other cost elements should also be investigated.

One other weakness, and they point to it themselves, is that all failures have the same probability. The reason for this simplification is that more probabilities exponentially increase the number of terms in the objective function, exponentially increasing the computational work that has to be done. This, they say, will be the object of further studies.

Secondly, the model assumes that all facilities are uncapacitated, which is, in most cases, an unrealistic assumption. Here too, further studies will be made to accommodate capacitated facilities.

Finally the authors note that the tradeoff may not be easily understood, since there is still is an element of uncertainty, since it assumes a certain probability for failure, which may or may not hold true. What matters to decison makers is hard numbers, not maybe numbers. On the other hand, visualizing tradeoff allows for a better understanding of the idea that improving reliability at a small cost can greatly reduce disruption costs.

Reference

Snyder, L., & Daskin, M. (2005). Reliability Models for Facility Location: The Expected Failure Cost Case Transportation Science, 39 (3), 400-416 DOI: 10.1287/trsc.1040.0107

Author links

• lehigh.edu: Lawrence V Snyder
• umich.edu: Mark S Daskin

Related

• husdal.com: Bad locations – bad logistics

Typical Norwegian

It is unfortunate that the decision-making process for infrastructure projects in Norway rests heavily on local and regional governments and less on the oversight of the national planning authorities. Overall socio-economic benefit is set aside to satisfy local desires or is simply overruled by successful local lobbyists, which is why you can find billion-dollar bridges or tunnels going to remote communities of a few hundred souls, while the roads in the big cities are heaving under the deadlock of congestion.

Building a new stretch of road or rail takes a lot of effort, a lot of preparation and a lot of organization. Up-sizing and down-sizing the efforts just for a few kilometers here and there is indeed a waste, and it would be better to have larger and longer-lasting projects.

Most transportation planners will agree on this, but for some reason Norway does not want to hand the decision of whether to build or not or how much to build or not over to the economists and experts, and rather let the politicians decide.

Does the term economies of scale ring any bells here? I say it does.

Links

• Dagbladet: Større vei- prosjekter gir mer vei

Related

• husdal.com: How come one of the world’s richest countries has one of the world’s worst road network?
• husdal.com: Cutting back on road spending may not be a wise thing to do

Monetized and Non-monetized impacts

The strength of the Norwegian method is how non-monetizable impacts are handled and integrated with the framework of the impact assessment. The significance of the various impacts is then assessed by combining the value and the magnitude of impact on five criteria: landscape/cityscape, community life and outdoor recreation, cultural heritage, natural environment and natural resources.

A valuable resource for transportation planners, the actual handbook is unfortunately only available in Norwegian.

Downloads

• vegvesen.no: Impact assessment in Norway (in English)
• vegvesen.no: Impact assessment in Norway (handbook, in Norwegian)

An alternative?

This book is rather heavy. If you’re looking for a lighter version, I suggest Cost-Benefit Analysis: Theory and Application by Tefvik Nas.

Reference

Boardman, A. et al. (2005). Cost-Benefit Analysis – Concepts and Practice. Prentice Hall.

Author link

• sauder.ubc.ca: Anthony Boardman

Husdal, J. (2007) Cost-Benefit Analysis – an essay about valuation problems. Unpublished working paper. Molde University College, Molde, Norway. Available at http://husdal.com/2007/01/05/cost-benefit-analysis-an-essay-about-valuation-problems/ Last accessed on [date].

Introduction

Cost-benefit analysis (CBA), in essence, is a tool for decision making. It can be applied to almost any kind of decision in any kind of field. In its most pure form, a CBA will aggregate the pros and cons (positive and negative effects) of a proposal, and, if the pros (benefits) outweigh the cons (costs), the proposal is viable. Usually, the analyst will assign monetary values to each of the costs and benefits, hence making the analysis easier to calculate, even if the cost and benefits per se are intangible, and thus, not directly expressible in money values. Problems often arise in how to assess the monetary values of both tangible and intangible effects, which may lead to skewed and biased results. One of the reasons for this is that most cost-benefit analyses are done so-called ex ante, before a project or proposal or policy is carried out or implemented.

The use of cost benefit analysis in the transportation sector

Cost-benefit analyses are widely used within the transportation sector. Albeit seemingly a new technique, one of the first to actually apply CBA was Dupuit, in France, in 1844, in his classic paper on the utility of public works (Prest, 1965). Since then, CBA has emerged as one of the most-used tools in deciding the viability of proposed infrastructure projects. A full CBA not only assesses the immediate impact and immediate costs and benefits (primary market effects), but also takes into account all externalities that are affected by said project (secondary market effects). Dupuit was also the first to introduce the concept of consumer surplus, a key element in economic welfare theory. Consumer surplus is defined as the difference between the maximum amount that an individual would be willing to pay for a good and the actual amount paid. On a standard supply and demand diagram, see below (taken from Wikipedia), consumer surplus is the triangle above the price and below the demand curve. These consumers are paying less for a good than the maximum that they would pay. Producer surplus shows up as a triangle below the price and above the supply curve, since that is the minimum price that a producer can produce that quantity with and still make a profit.

Consumer surplusCombined, consumer surplus and producer surplus make up what is called social surplus, and a cost-benefit analysis seeks to identify the social surplus of the particular project that is subject to analysis. A cost-benefit analysis proceeds in four essential steps: (a) identification of relevant costs and benefits, (b) measurement of costs and benefits, (c) comparison of cost and benefit streams accruing during the lifetime of a project, and (d) project selection. (Nas, 1996). In the first phase, all related costs and benefits are identified. The second stage then, entails valuing and pricing of both tangibles and intangibles. In the third phase, the present value of future benefits and costs must be calculated, and finally, projects are ranked according to some criteria, most often cost-benefit ratio and net present value. However, choosing which project to go forward with may not be straightforward. Sometimes, external effects may play a part in the final decision, and often there will also be a political or regional agenda that needs to be satisfied.

In Norway, the Planning and Building Act requires the use of an environmental impact analysis for all major development projects. Albeit named “impact analysis”, in essence it is a full CBA, assessing both tangible and less tangible effects (Statens Vegvesen, 2006).

Reliability and vulnerability = benefit and cost?

One issue of major concern in the transportation sector, which has gained interest recently, is the reliability of infrastructure systems. Road transportation is no exception, since road networks are the main backbone of modern society. Consequently, the reliability, or conversely, the vulnerability of any transportation network is thus a decisive factor not only in terms of market outreach and competition, but also in terms of continuity, to ensure a 24/7 operation of the community we live in. Any threat to the reliability of the transportation network constitutes a vulnerable spot, a weakness. This is of particular concern when considering sparse, rural networks, because what by urban standards is a minor degradation (i.e. car accident, resulting in queuing, delays and diversions) may have severe consequences if occurring in a rural setting (i.e. blocking the only access road for hours).

In the most common manner of speaking, the reliability of a transportation network can be defined as the probability that one or more of its links functions, or rather: does not fail to function, according to a set standard of operating variables. A non-functioning, or at best, badly-functioning link will impose costs on the user in terms of loss of time, additional operation costs or other costs as a result of delays and diversions. Transporters of perishable goods will also experience a loss of value. Few will question that the sender, the recipient, the freight hauler, or society at large, experience additional costs when goods or people cannot reach their destinations in time or in space (Husdal, 2004).

The increase in just-in-time (JIT) manufacturing operations has made a reliable travel time an important economic factor. JIT relies on the transportation system to take advantage of lowcost labor and manufacturing plant development costs. Producing components in several manufacturing plants and bringing them together in one location at the same time to produce the final product can reduce inventory requirements and total costs, but requires a controlled environment for travel times. If one component does not arrive due to improper product scheduling or due to traffic delays, an assembly line can be shut down, or costly building space has to be used for inventory storage, rather than for manufacturing or assembly operations (Lomax, 2003). Consequently, it should be obvious that a reliable transportation network represents a benefit to society. Equally, a vulnerable network would represent a net cost to society. These are socio-economic costs that should be taken into consideration in cost-benefit analyses of transportation projects; yet, as a norm, reliability and vulnerability are not evaluated in today’s practice (RISIT, 2004). This means that the evaluation of benefits of proposed new projects and the costs of using existing infrastructure may be lacking important elements.

In terms of road user utility new projects are generally considered as improvements, the same normally applies to reliability. It is often taken for granted that road improvements that alleviate congestion and travel time variability also improve reliability. What remains an open-ended question is whether new projects also bring with them a lesser vulnerability. Exchanging one kind of vulnerability for the other, while supposedly increasing reliability, may in terms of socio-economic impact not be the best solution.

A system approach that takes both sides into account is thus necessary. The reason is straightforward: to evaluate the cost of remaining vulnerable against the assumed benefit of becoming less vulnerable with the proposed project. Vulnerability, represented by the consequential costs of an operational degradation, is a cost that should be included in a cost-benefit analysis. Likewise, reliability, representing the consequential benefits of an operational improvement, is a benefit value, which too needs to be included (Husdal, 2005).

Example 1: A project is proposed to reduce the consequences of avalanches on a selected link on a given route, being one route of several alternative routes between a given origin and destination. This reduces the particular vulnerability of this link on this route. This improvement may cause motorists to transfer to this route from the alternative routes. If some remaining vulnerabilities on other (non-improved) links along the now partially improved route are neglected and left unattended, then this may in turn affect a larger number of road users then before the particular improvement.

Example 2: Typical Norwegian issues of vulnerable links are ferry crossings, with a high probability of frequent closures and delays, often being the only access to a given community or between major regions. These crossings are in some cases replaced by sub sea tunnels, with practically no probability of any closure, and are hence considered a considerable improvement in reliability. Normally, after the tunnel is built, the ferry jetties are dismantled. If the tunnel is then closed due to an accident, say, a fire, and remains closed for a long period of time, this community would in fact be more vulnerable after the project then before. Disruptions in a road network are often the result of external influences such as landslides, avalanches or rock fall, flooding, lack of snow clearing during winter, accidents that require extensive clean-up, etc.; the list of scenarios is literally endless. Intentional sabotage or terrorist attacks must also be included here, but are not a major scope of this paper. Often seen in Norway, closures of sub sea tunnels and failures or severe degradation of ferry services are also among the many factors that will occur depending on the season and location. Ferry services may be of no concern in some parts of the country, whereas rock fall can be a major issue in other parts.

Figure 2 below, taken from Husdal (2005),  is a good illustration of the relationship of cost-benefit and vulnerability versus reliability: The disruptions costs, and thus vulnerability, increase from right to left (solid line), the cost of countermeasures to overcome potential disruptions, and hence the assumed reliability, increase from left to right (dotted line).

Example 3: To keep a mountain road open during winter one may consider improving the road by building a tunnel, more snow sheds, or aligning the road differently to alleviate weather exposure, or alternatively, invest in better snow-clearing equipment and increase clearing frequency. Today’s situation is (A), with high disruption costs. Building a new road and/or a tunnel is costly (B), but the probability of disruptions is lessened considerably and the disruption costs are almost negligible. Better snow-clearing equipment (C) reduces the susceptibility to disruptions somewhat – the road is still exposed to severe weather conditions – but the investment costs are much lower than in the former alternative. Consider which alternative is the most beneficial to society if investment costs and saved disruption cost are weighed against each other? Note point (D) where the two curves intersect. The expected cost of disruptions (or the expected benefits of avoiding disruptions) is there equal to the cost of countermeasures. D is the socio-economically optimal level of expected disruption costs (and the optimal expenditure on countermeasures). A movement towards D from the left means that it will be cost-effective to implement a countermeasure. A movement from D to the right means that society will be better off “living with the vulnerability”. In this particular case, better snow-clearing equipment will be a good countermeasure, a new road/tunnel will not. The individual road user may of course disagree here, but from an overall cost-benefit perspective it is correct.

If interpreting the curves in figure 2 in an economic setting, it could be said that the marginal cost of disruptions initially will fall sharply while the marginal cost of countermeasures will only rise slightly. The more countermeasures that are put in place, the less extra benefit is achieved for each marginal investment in reliability. This is in line with the traditional way of thinking in contingency planning, that it often only takes small changes or investments to make a considerable impact. Full and 100% reliability however is very costly, due to the unpredictable manner of the potential disruptions. From a strictly economical point of view, the cost of increasing the reliability should not exceed the cost of vulnerability for society to experience a benefit. What is apparent is that it may be straightforward to quantify the investment costs associated with an effort to increase reliability, the costs of disruptions are much harder to quantify in measurable terms, albeit it can be done.

Another way to look at this figure is to see the two graphs as supply/demand curves, where the cost of countermeasures corresponds to the cost of supplying a certain level of reliability. The cost of vulnerabilities can be seen as demand for the level of reliability represented by the countermeasure curve.

Valuation of reliability and vulnerability

If we see reliability as a benefit and vulnerability as a cost, a valuation of reliability and vulnerability would then entail weighing the costs of disruptions against the cost of avoiding these disruptions. Going back to the illustration of JIT in chapter 3, a transportation-dependent manufacturer will seek to either establish measures to counter disruptions that occur, or simply be willing to bear the additional costs that a disruption may cause. If a manufacturer is dependent on ontime his input materials, he may choose to heighten his input inventory, just in case his supply is cut off. If on-time delivery of his output is the issue, a manufacturer may choose to produce multiple sets of the same product, which could be shipped using different routes or modes of transportation, in case one route is cut off, or he might give his customers a rebate for deliveries that are delayed. The costs of these measures are, naturally enough, would have to be passed on to the customers.

The customers then face a similar dilemma, should one be willing to pay the extra premium for on-time delivery with one manufacturer or choose a competitor with a lower price and bear the cost of a possible delayed delivery? The economic implications are clear, but do these problems actually constitute a socioeconomic cost, or is it just a transfer of costs and benefits? If one manufacturer looses a contract because he cannot deliver on time, while another one, who would not have gotten a contract in the first place, gains said contract, for society as a whole that makes no difference. It could, however, have an impact on the regional economy, i.e. the secondary market.

Furthermore, since transportation networks, e.g. roads are supplied by the government, if disruption losses are directly inferable from a lack of adequate road structures, and that is in fact more often than not the case, then the benefits of reliability and the costs of vulnerability indeed are issues that need to be addressed for the society as a whole. Having come this far, and subscribing to the view that investments in reliability constitute a cost and saved disruption costs constitute a benefit, now remains the issue of finding the measure that best expresses these costs and benefits in monetary values.

Usually, generalized travel cost is the driving factor in cost-benefit analyses in the transportation sector. In this particular setting, savings in disruption costs correspond to savings in additional travel costs that occur as a result of detours, delays and/or closures. Normally, these costs should be easy and straightforward to calculate. An example of how this can be done on a large scale can be found in Jenelius et al. (2006). Here, the importance and exposure to disruption of each road link in the road network in Northern Sweden is characterized by how the particular link contributes to the increase in travel time, and hence travel cost, for all origin-destination (O-D) pairs within the network. For a given O-D pair it is then possible to calculate the potential vulnerability cost of travelling origin and destination should a given link in the network fail. The cost of improving reliability is not so straightforward to calculate, since that will depend on the nature of the disruption and how it best can be alleviated. Nonethesless, using Jelenius’ methodology it is possible to find (a) the vulnerability cost of using a particular route (or link on a route) in a transportation network and also (b) the vulnerability cost of location in relation to the neighbouring transportation network.

One application that comes to this author’s mind is avalanche and rock fall protection. In the mid-nineties (to be updated this year) the Norwegian Public Roads Administration surveyed “all” potential hazard sites in the region of Møre og Romsdal, registering frequency of occurrences and past closure times and suggesting abatements, including how much these would cost. However, the work stopped here, most likely since although the cost-benefit ratio for each site could be calculated, ranking the sites in a sensible manner may have proven too difficult, not to mention probable budget constraints. In this case, one could have used Jenelius et al. (1996) to calculate the disruption costs for all O-D pairs, combine this with the probability of failure and the duration thereof, and then matched with the sum of the abatement costs for all O-D pairs to suggest where to focus the abatement efforts, by finding the sites with the overall highest benefit/cost-ratio.

Generalized travel cost or saved travel time, or more specific, the value of saved travel time is normally found by using contingent value (CV) studies, in which respondents are asked how much they would be willing to pay (WTP) for a certain gain in travel time, or how much they would be willing to pay to accept (WTA) a certain increase in travel time. It should be noted though that the willingness to pay for a certain gain often does not corresponds to the willingness to accept the same amount of loss (Coursey 1987), in that the willingness to accept, or compensation demanded, often far exceeds the price one is willing to pay for a gain of the same size as the loss. What this means is that people tend to underperceive the value of losses and overperceive the value of gains. Nevertheless, CV studies have an important place in assessing the value of costs (compensation demanded) and benefits (gain received). Within the realm of reliability and vulnerability in the transportation sector it thus should be investigated in what manner transport-dependent entities adapt to transport-related uncertainties, and (a) what cost they choose to bear for this adaptation (which is essentially their WTA), and (b) what price they would be willing to pay (their WTP) to save these costs.

Conclusion

Reliability and vulnerability are important elements in ensuring a smooth 24/7 operation of transportation networks, road networks in particular, and cost-benefit analyses of transportation projects should take this into account. One way of doing this is to say that investments in improved reliability constitute a cost and thereby saved disruption costs constitute a benefit. To make a project viable in a cost-benefit sense the net present value should be positive. In this case: the present value of future disruption costs should outweigh the presently needed investment to achieve the future return. Both gains and losses in travel cost and investment cost need to be assessed carefully, but if done properly, it should be possible to develop a vulnerability index for a road network that is able to assign to both usage of and location on a particular part of the network.

References

1. Boardman, A.E. et al., 2006, Cost-benefit analysis: concepts and practice, Pearson Prentice Hall

2. Coursey, D. L. et al., 1987, The Disparity Between Willingness to Accept and Willingness to Pay -Measures of Value, The Quarterly Journal of Economics, Vol. 102, No. 3, pp. 679-690

3. Husdal, J., 2004, Pålitelighet og sårbarhet – et ikke-tema i nyttekostanalyser? Samferdsel 2/2004, s. 28-30. Reliability and vulnerability – a non-issue in cost-benefit analyses? Samferdsel (Journal of the Norwegian Institute for Transport Economics, TØI), 2/2004, pp. 28-30

4. Husdal, J, 2005, The vulnerability of road networks in a cost-benefit perspective. Proceedings of the the Transportation Research Board Annual Meeting 2005, Washington DC, USA, 9-13 January 2005

5. Jenelius, E. et al., 2006, Importance and exposure in road network vulnerability analysis, Transportation Research Part A: Policy and Practice, Vo. 40, Iss. 7, pp. 537-560

6. Lomax, T., et al., 2003, Selecting Travel Reliability Measures, Texas Transportation Institute and Cambridge Systematics Inc.

7. Prest, A. R., Turvey, R., 1965, Cost-Benefit Analysis: A Survey, The Economic Journal, Vol. 75, No. 300, pp. 683-735

8. Nas, T.F., 1996, Cost-benefit analysis: Theory and application, Sage publications

9. RISIT, 2004, RISIT – Risk and safety in the transport sector – A state-of-the-art review of current knowledge. White paper. The Research Council of Norway.

10. Statens vegvesen, (2006) Konsekvensanalyser: Del 1, Prinsipper og metodegrunnlag. Håndbok 140. Vegdirektoratet. (Norwegian Public Roads Administration (2006) Impact analysis: Principles and methodology. Handbook 140.  Oslo, Norway.)

READ ONLINE

See the presentation

INTRODUCTION

Few will question that the sender, the recipient, the freight hauler, or society at large, experience additional costs when goods or people cannot reach their destinations in time or in space. Consequently, it should be obvious that a reliable transportation network represents a benefit to society. Equally, a vulnerable network would represent a net cost to society (1). Why then, is the reliability, or conversely, the vulnerability, of the transportation network not a matter of evaluation in traditional cost-benefit analyses? In the most common manner of speaking, the reliability of a transportation network can be defined as the probability that one or more of its links functions, or rather: does not fail to function, according to a set standard of operating variables. A non-functioning, or at best, badly-functioning link will impose costs on the user in terms of loss of time, additional operation costs or other costs as a result of delays and diversions. Transporters of perishable goods will also experience a loss of value. These are socio-economic costs that should be taken into consideration in cost-benefit analyses of transportation projects; yet, as a norm, reliability and vulnerability are not evaluated in today’s practice (2). This means that the evaluation of benefits of proposed new projects and the costs of using existing infrastructure may be lacking important elements.

Taking up the invitation in Berdica (3) to bring out and recognize the vulnerability in the road transportation system as a meeting point for all the different strands of transportation reliability research and other issues (p.127), the focal point of this paper is to look at a road network from a reliability and vulnerability perspective and to link this analysis to cost-benefit decisions. Looking beyond the science of vulnerability assessments, this paper discusses some of the network attributes that influence the vulnerability of transportation networks, influences that can be described as structure-related, nature-related or traffic-related attributes. The paper introduces vulnerability as a parameter for decision-support in cost-benefit analyses and seeks to establish a link between the terms reliability and vulnerability vis-à-vis costs and benefits.

First, the paper exemplifies why vulnerability should be an issue in road project evaluations or in determining public service locations. Second, it considers how the study of the vulnerability of transportation networks relates to traditional reliability analysis. Third, potential measures of vulnerability are briefly described and classified into three separate groups. Fourth, an attempt is made to establish a link between the terms reliability and vulnerability and the terms costs and benefits.

WHY MAKE VULNERABILITY AN ISSUE?

Transportation networks like freeways and interstate highways are the main backbone of modern society. Consequently, the reliability, or conversely, the vulnerability of any transportation network is thus a decisive factor not only in terms of market outreach and competition, but also in terms of continuity, to ensure a 24/7 operation of the community we live in. Any threat to the reliability of the transportation network constitutes a vulnerable spot, a weakness (4)(5).

This is of particular concern when considering sparse, rural networks, because what by urban standards is a minor degradation (i.e. car accident, resulting in queuing, delays and diversions) may have severe consequences if occurring in a rural setting (i.e. blocking the only access road for hours). From a user point of view, what matters most in relation to a transportation network is the following: Can I, at the desired time of departure, get from A to B by using the intended route and means of transportation, and arrive at a desired time, which would be the best case. Or, does there exist no route or means of travel at all that can take me from A to B at the desired time of departure, let alone within arriving at the desired time, which to the user would be the worst case.

Disruptions in a road network are often the result of external influences such as landslides, avalanches or rock fall, flooding, lack of snow clearing during winter, accidents that require extensive clean-up, etc.; the list of scenarios is literally endless. Intentional sabotage or terrorist attacks must also be included here, but are not a major scope of this paper. Often seen in Norway, closures of sub sea tunnels and failures or severe degradation of ferry services are also among the many factors that will occur depending on the season and location. Ferry services may be of no concern in some parts of the country, whereas rock fall can be a major issue in other parts. In the most common manner of speaking, the reliability of a transportation network can be defined as the probability that one or more of its links functions, or in better said, the probability that one or more of its links does not fail to function. Vulnerability, on the other hand, represents the network’s or the links’ susceptibility to failure, where the term failure here expresses a considerable deviation from the normal functioning state of the link or network.

A non-functioning, or at best, badly-functioning link will impose costs on the user in terms of loss of time, additional operation costs or other costs as a result of delays and diversions. Transporters of perishable goods will also experience a loss of value. These are socio-economic costs that should be taken into consideration in cost-benefit analyses of transportation projects; yet, as a norm, reliability and vulnerability are not evaluated in today’s practice. New transportation projects are seldom explicitly evaluated in terms of increased reliability or lessened vulnerability (6). Nonetheless, vulnerability assessments should be acknowledged as an integral element within the cost-benefit analysis that is part of any transportation planning process.

A perhaps myopic, but at the same time illustrative case from Norway may demonstrate why vulnerability should be an issue in the evaluation of transportation networks:

On the North-western coast of Norway (figure 1), there is a hospital in the cities of Molde and Kristiansund, 75 kilometers (47 miles) apart. In conjunction with a recent evaluation of whether one should maintain two separate hospitals versus establishing one hospital at a new location that would cover the service areas of both hospitals, a study was conducted to compare the travel times for employees, out-patients and in-patients for a set of locations. What is disturbing to this author, and which is the reason for his interest in the subject, is that the study exclusively used “ideal” travel time parameters, among which: Speed equal to speed limit, and a waiting time of maximum 15 minutes at the required ferry crossings. What from the point of this author is lacking is a consideration of a) seasonal variations of serviceability (snowy and icy roads versus clear roads), b) the possibility of ferry breakdowns due to engine failure, ferry delays or cancellations due to high winds and rough seas (a well-known event in coastal Norway), c) potential closures of the mountain passes that must be traversed in this example (a season-dependent and location-dependent issue), d) possible closure of a sub sea tunnel for certain travelers, e) closure of long bridge spans due to high winds, and f) avalanches and rock fall, to name but a few. The aforementioned are all known incidents along the studied routes. However, these factors were not considered. Admittedly, incorporating these factors is difficult, and in a statistical sense perhaps of no significance. Furthermore, ideal parameters are, after all, easier to use for a ceteris paribus comparison of travel times. Nevertheless, should the vulnerability of the transportation network not be part of the equation when locating critical public services such as hospitals? It is not the point here to advocate the use of worst-case scenarios; however, they should be part of the overall evaluation. In hindsight, it is usually better to have evaluated worst-case scenarios and discarded them if they do not have any influence, than actually to have to face a worst-case scenario and wonder why this had not been evaluated.

With this generally applicable example in mind, it may be surprising that the reliability of the transportation network still remains a factor that is only seldom looked at in economic impact assessments. For the most part, saved travel time is what drives the policy of the decision makers, increased reliability is as a rule not a subject for closer evaluation; it is mostly taken for granted as far as new projects are concerned. Introducing the reliability of the road network as a decision variable will thus bring new perspectives to cost-benefit analyses, where the question of whether travelers and goods indeed can traverse the network and thus arrive at their destination is given more weight than in traditional analyses. From a freight hauler’s point of view, a vulnerable network is a network that is easily disrupted, resulting in unpredictable stops and downtime. This is probably seen as a much larger problem than a congested and slow-moving network that is relatively reliable and stable. In the latter, there is at least some guarantee that the goods will arrive at their destination, and most important, the transportation costs, though annoyingly inefficient, are still calculable and lead times are still predictable.

RELIABILITY VERSUS VULNERABILITY

It seems fair to conclude that the current literature acknowledges reliability and vulnerability as being two related, yet different concepts. Both Berdica (3) and Taylor and D’Este (7)(8) lament the lack of a formal definition of vulnerability. Traditionally, when looking at the reliability of transportation networks, it is done with a systems engineering approach. Reliability is here an expression of the probability that links within the network will function, reliability may thus be viewed as the degree of stability of the quality of service that a system offers.

In Bell and Iida (9), transportation network reliability is focused on connectivity reliability (also named terminal reliability) and travel time reliability. Nicholson et al. (10) in addition list and discuss encountered reliability, capacity reliability and flow decrement reliability as ways of measuring reliability. Whereas probability or predictability is a major concern in network reliability studies, the impacts or consequences of disruptions are the main focus of vulnerability studies (11).

Vulnerability and reliability are two related concepts (7), but network vulnerability relates to network weaknesses and the economic and social consequences of network failure, not so much the probability of failure. In (3), vulnerability is related to serviceability, namely the possibility to use a link, route or road in a network at a given time. Vulnerability, then, is the inability to supply adequate serviceability and serviceability as such is determined by a set of performance measures.

In (8), vulnerability relates to the degree of accessibility of a given node in the network, where accessibility is expressed as the travel cost needed to access the particular node, comparing optimal and alternative routes or detours. LLeras-Echeverri and Sanchez-Silva (12) extend the terminal reliability as seen in (9), to include a progressive failure scenario approach to identify weak links.

What is apparent then, is that vulnerability, unlike reliability, needs to be more than a quantitative probability calculation related to the functioning or non-functioning of a network link. Transportation networks are vulnerable to a wide range of possible scenarios, events and incidents, some probable, some improbable, some disastrous, some with only minor disturbances.

Reliability focuses on the possibility of maintaining a link during these scenarios; vulnerability focuses on the possibility of disrupting or degrading a link or network. To be of use in a vulnerability setting, the term operability needs to be approached differently for different types of roads, road conditions, goods or transports. If associating reliable with being operable, and associating vulnerable with being non-operable, similar to serviceability in (3), then it is warranted to juxtapose reliability with vulnerability.

Reliability, in this sense, means non-vulnerability or exhibiting a high degree of operability under any circumstances; vulnerability means non-reliability or exhibiting a low degree of operability under certain circumstances. The following definitions can be set up to distinguish and juxtapose reliability and vulnerability:

1. Reliability describes the operability of the network under varying strenuous conditions (i.e. the ability to continue to function).

2. Vulnerability describes the non-operability of the network under varying strenuous conditions (i.e. the susceptibility to fail to function).

3. A reliable network exhibits a high degree of operability as expressed by its serviceability, accessibility, and non-variability under any circumstance, due to the presence of redundancy, robustness, and resilience in the network.

4. A vulnerable network exhibits a low degree of operability as expressed by non-serviceability, non-accessibility, and variability under certain circumstances, due to the lack of redundancy, robustness, and resilience in the network.

5. Vulnerability = Non-Reliability (under said certain circumstances)

This is how the terms vulnerability and reliability are used throughout this paper. Vulnerability can thus be defined as the consequential cost of a lack of reliability under certain circumstances, and this consequential cost must comprise not only the immediate toll on the road-users, but the overall socio-economic costs on the community that this vulnerability would entail.

MEASURES OF VULNERABILITY

Transportation networks are susceptible to a wide range of vulnerabilities that can lead to an operational degradation. These vulnerabilities are the result of certain attributes or qualities pertaining to the network itself (13), and one way of categorizing attributes and influences is to sort them in terms of structure, nature and traffic (14).

Structure-related or structure-generated vulnerability pertains to the way the road is built, and to attributes of the road network itself, not only in terms of network topology and connectivity, but also in terms of the physical body of the road, geometry, width, curvature, gradient, tunnels, bridges, weight restrictions, etc.

Nature-related or nature-generated vulnerability pertains to attributes of the natural environment, the topography and the terrain that the road traverses, and to natural incidents, such as flash floods, avalanches, rock fall, snow and ice, fog, earthquakes, tsunamis, climate change, and so on.

Traffic-related or traffic-generated vulnerability pertains to attributes describing the generic flow of traffic and attributes resulting in flow decrements, such as daily rush hour and weekend highs, as well as maintenance operations, snow clearing, accident clear-up, and ongoing construction works.

Typically, these vulnerabilities will occur on a collective basis, rather than on a one-by-one basis. Even though a particular stretch of road may be susceptible to only one of the aforementioned vulnerabilities, the overall network will be exposed to the full set of all vulnerabilities, some acerbating the other (figure 2 above). It is the collective sum of these vulnerabilities that needs to be addressed. Some links may exhibit structural deficiencies, some will be at the mercy of nature, and others are particularly vulnerable to traffic-generated incidents. Hence, a vulnerability analysis must consider each attribute separately, and, at the same time, as a whole.

In addition to the three categories listed above, a fourth dimension should be mentioned: the vulnerability towards an intentional terrorist attack, since an attacker will seek to exploit the vulnerabilities that are present in the network.

In essence, a vulnerability analysis must answer the three following questions, but not necessarily in the order mentioned: First, Vulnerable…where? to assess the location. Second, Vulnerable…to what? to assess the particular circumstances. Third, Vulnerable…how? to address the particular scenario and its impact.

VULNERABILITY AND RISK

Vulnerability and risk and are related concepts (15). An often used definition of risk is risk as being the product of consequence and probability. This equation can be extended to include vulnerability, such that vulnerability is as a factor that contributes to lessening or increasing the risk:

R = V(ec) x P(ec)

V = The vulnerability to the occurrence of an external circumstance (ec) or threat
P = The probability of the occurrence of an external circumstance (ec) or threat

One important point with respect to terrorist threats is that, while other threats can be expected to occur in a somewhat random and stochastically predictable fashion, terrorist threats are less predictable and somewhat endogenous (in the sense that the threat is likely to be greater if the vulnerability is greater, whereas with natural threats the probability of threat and the vulnerability to the threat can be assumed to be independent of one another). In terms of risk management, one usually distinguishes between two distinct strategies, one aimed at reducing the probability of an incident occurring and the second aimed at reducing the consequences of an incident that occurs, reflecting back towards the earlier definition of risk.

Applying this notion to the reliability and vulnerability of a road network, the aforementioned strategies would be aimed at a) reducing the probability of a disruption, known as pre-emptive measures, and b) reducing the consequences of a disruption, known as mitigative measures.

VULNERABILITY/RELIABILITY AND COST/BENEFIT

With the subject of vulnerability being open to a wide range of research approaches, it is not surprising that the vulnerability of transportation networks lacks a consensus definition. What should not be dissented from though, is that any vulnerability carries with it a cost to society, not only to the road users. The purpose of a cost-benefit analysis is to weigh the costs of a proposed project against the benefits of the project (16)(17)(18). If the benefits exceed the costs, then the project increases society’s welfare. If the costs exceed the benefits, society will experience a loss of welfare. The argument of vulnerability versus reliability is analogous: Any vulnerability of the transportation system causes disruptions that cause costs, which are a loss of welfare; vulnerability is thus a cost that is quantifiable. If society puts measures in place to reduce the vulnerability of the transportation network, the increased reliability represents a benefit.

In terms of road user utility new projects are generally considered as improvements, the same normally applies to reliability. It is often taken for granted that road improvements that alleviate congestion and travel time variability also improve reliability. What remains an open-ended question is whether new projects also bring with them a lesser vulnerability. Exchanging one kind of vulnerability for the other, while supposedly increasing reliability, may in terms of socio-economic impact not be the best solution.

A system approach that takes both sides into account is thus necessary. The reason is straightforward: to evaluate the cost of remaining vulnerable against the assumed benefit of becoming less vulnerable with the proposed project. Vulnerability, represented by the consequential costs of an operational degradation, is a cost that should be included in a cost-benefit analysis. Likewise, reliability, representing the consequential benefits of an operational improvement, is a benefit value, which too needs to be included. Today’s project evaluation practice looks almost exclusively at the construction costs on one hand, and the road user utility as the benefit on the other hand, with little or no notion of reliability or vulnerability.

Example 1: A project is proposed to reduce the consequences of avalanches on a selected link on a given route, being one route of several alternative routes between a given origin and destination. This reduces the particular vulnerability of this link on this route. This improvement may cause motorists to transfer to this route from the alternative routes. If some remaining vulnerabilities on other (non-improved) links along the now partially improved route are neglected and left unattended, then this may in turn affect a larger number of road users then before the particular improvement.

Example 2: Typical Norwegian issues of vulnerable links are ferry crossings, with a high probability of frequent closures and delays, often being the only access to a given community or between major regions. These crossings are in some cases replaced by sub sea tunnels, with practically no probability of any closure, and are hence considered a considerable improvement in reliability. Normally, after the tunnel is built, the ferry jetties are dismantled. If the tunnel is then closed due to an accident, say, a fire, and remains closed for a long period of time, this community would in fact be more vulnerable after the project then before.

Figure 3 illustrates the relationship between vulnerability and reliability. The investment costs, and hence the assumed reliability, and thus, increase from left to right (dotted line), disruptions costs, and thus vulnerability, increase from right to left (solid line). From a strictly economical point of view, the cost of increasing the reliability should not exceed the cost of vulnerability for society to experience a benefit.

What is apparent is that it may be straightforward to quantify the investment costs associated with an effort to increase reliability, the costs of disruptions are much harder to quantify in measurable terms, albeit it can be done. The investment costs are included as a cost in cost-benefit analyses; however, saved disruption costs are not included as a benefit. Consequently, a proposed investment is not valued correctly, if saved disruption costs are not properly accounted for. The question that arises is whether saved disruption costs represent a utility similar to saved travel costs.

Example 3: To keep a mountain road open during winter one may consider improving the road by building a tunnel, more snow sheds, or aligning the road differently to alleviate weather exposure, or alternatively, invest in better snow-clearing equipment and increase clearing frequency. Today’s situation is (A), with high disruption costs. Building a new road and/or a tunnel is costly (B), but the probability of disruptions is lessened considerably and the disruption costs are almost negligible. Better snow-clearing equipment (C) reduces the susceptibility to disruptions somewhat – the road is still exposed to severe weather conditions – but the investment costs are much lower than in the former alternative. Consider which alternative is the most beneficial to society if investment costs and saved disruption cost are weighed against each other?

Note point (D) where the two curves intersect. The expected cost of disruptions (or the expected benefits of avoiding disruptions) is there equal to the cost of countermeasures. D is the socio-economically optimal level of expected disruption costs (and the optimal expenditure on countermeasures). A movement towards D from the left means that it will be cost-effective to implement a countermeasure. A movement from D to the right means that society will be better off “living with the vulnerability”. In this particular case, better snow-clearing equipment will be a good countermeasure, a new road/tunnel will not. The individual road user may of course disagree here, but from an overall cost-benefit perspective it is correct.

VULNERABILITY AND MULTI-CRITERIA ANALYSIS

According to (6), the transportation sector in general has very limited experience with regard to risk based management, and that especially thought processes on acceptable risk and the use of acceptance criteria is not fully developed. Cost-benefit analyses and environmental impact analyses are being used, but risk analyses and risk acceptance criteria are not really used. Risk as a concept and as a management tool has no marked tradition among the Norwegian road authorities or even amongst the international road authorities. With vulnerability inextricably linked to risk, a similar argument can be made with regard to vulnerability based management.

Some of the elements that project evaluation procedures should take explicitly into account in order to incorporate considerations of vulnerability are the following:

1. The probability and impact of failure of a given network, link or route, given various external circumstances or strenuous conditions

2. The probability of those external circumstances occurring

3. The robustness of the system (i.e., the probability that the system will continue to function even if a threat eventuates at a vulnerable point)

4. How long it will take (and how expensive it will be) to repair the system if the threat occurs and the system fails at its vulnerable point

5. What the costs are to the general economy of such a failure (i.e., goods and passengers not getting to their destinations, or getting there late, transportation carriers being forced to use expensive detours, etc.)

6. The contribution of a given project to improving the robustness (and hence reliability) of the system

7. What degree of risk aversion that should be applied in deciding what weight to place on the risk (i.e., level of threat times vulnerability) that has been identified.

These considerations can be incorporated into the project evaluation process using a multi-criteria analysis for both monetary and non-monetary impacts.

Figure 4 illustrates how this can be done. Here, vulnerabilities are assessed by their effect or impact. The purpose of dividing vulnerabilities into individual effects or impacts is to assess which type of traffic that is most affected or which impacts that are most prominent. The figure shows some of the effects of a sudden and unforeseeable incident that leads to the closure of a link or route (19).

Note that the evaluation criteria and the increment values used in figure 4 are for illustration only. In this illustration, average annual daily traffic serves as the first criteria, to assess the number of users that is affected. Second, the waiting time or delay time, or the time it takes to drive the diversion is differentiated between light and heavy vehicles to roughly account for the difference in value of time for business and leisure trips. Third, perishable goods lose value if not delivered on time. Fourth, some business may experience a loss of profit, because important deliveries or shipments are not made. Fifth, combined with the other criteria, the time (and cost) of reconstruction of this particular link serves as an additional criteria. The number of evaluation criteria must be adapted to the scenario that is investigated. Evaluation categories A-Z are weighted on a 0-1 scale, such that the sum of the category weights equals 1. Each weight is multiplied with the impact value 1-5. With this approach it is possible to compare and evaluate a number of vulnerabilities and to find the most critical link or most vulnerable point according to the criteria used.

CONCLUSION

This paper has sought to bring together three fields of research: reliability and vulnerability, costs and benefits, and decision-making, with the rationale that the reliability or vulnerability of the transportation network only seldom is looked at as a decision parameter in cost-benefit analyses, especially for new projects.

This paper has advocated that reliability and vulnerability should be part of the project evaluation in road development projects, and that reliability and vulnerability should be accounted for in cost-benefit analyses of said projects. A solid decision-making process should include a cost-benefit evaluation, but decisions related to vulnerability must pass a scrupulous review before they can be executed. Presentations of consequences should serve to highlight the issues that are at stake; these should not merely be synthesized into measures of gains and losses (20).

Socio-economic calculations and multi-criteria evaluations can assist in addressing the individual effects of vulnerability and reliability in the road network. In the end, the evaluation criteria, the factors, the constraints, the individual weighting and the decision rules are at the discretion of the decision-maker, and by including reliability and vulnerability as one element in the decision-making process it is possible to evaluate the cost of remaining vulnerable against the assumed benefit of becoming less vulnerable with the proposed project.

Finally, it must be said that the aim of this paper was not to find a justifiable measure of vulnerability, but to point at vulnerability as a variable that needs to be taken account of in cost-benefit analyses and in the evaluation of transportation networks. Therefore, in its composition, this paper has likely raised more questions than provided solid answers. It remains to bee seen, then, whether vulnerability and reliability will be a driving force in cost-benefit analyses of future road development projects.

REFERENCES

1. Husdal, J. (2004) Pålitelighet og sårbarhet – et ikke-tema i nyttekostanalyser? Samferdsel 2/2004, s. 28-30. Reliability and vulnerability – a non-issue in cost-benefit analyses? Samferdsel (Journal of the Norwegian Institute for Transport Economics, TØI), 2/2004, pp. 28-30

2. Statens vegvesen (1995) Konsekvensanalyser: Del 1, Prinsipper og metodegrunnlag. Håndbok 140. Vegdirektoratet. (Norwegian Public Roads Administration (1995) Impact analysis: Principles and methodology. Handbook 140.), Oslo, Norway.

3. Berdica, K. (2002) An introduction to road vulnerability: what has been done, is done and should be done. Transport Policy, Vol. 9, Iss. 2, pp. 117-127

4. Justisdepartementet (2001) Transportsikkerhet, i: NOU 2000:24 Et sårbart samfunn. (Norwegian Ministry of Justice (2001) Transportation Safety, in: NOU (Government Report) 2000:24 A vulnerable society.)

5. Hagen, J.M.; Rodal, G.H.; Hoff, E.; Lia, B.; Torp, J.E.; Gulichsen, S (2003) Beskyttelse av samfunnet med fokus på transportsektoren. FFI/Rapport 2003/00929. Forsvarets Forskningsinstitutt, Kjeller, Norge. Protecting the society, focusing on the transport system. FFI/Report 2003/00929. Norwegian Defence Research Establishment, Kjeller, Norway.

6. RISIT (2004) RISIT – Risk and safety in the transport sector – A state-of-the-art review of current knowledge. White paper. The Research Council of Norway.Adler, H.A. (1987) Economic Appraisal of Transport Projects. Johns Hopkins University Press.

7. Taylor, M. A. P. and D’Este, G.M.D. (2003a) Network Vulnerability: An Approach to Reliability Analysis at the Level of National Strategic Transport Networks. In: The Network Reliability of Transport, eds. M.G.H. Bell and Y. Iida, Proceedings of the 1st International Symposium on Transportation Network Reliability, pp.23-44. Elsevier Science, Kidlington.

8. Taylor, M. A. P. and D’Este, G.M.D. (2003b) Concepts of network vulnerability and applications to the identification of critical elements of transport infrastructure. Paper presented at the 26th Australasian Transport Research Forum, Wellington, New Zealand, 1-3 October

9. Bell, M.G.H., Iida, Y. (1997) Network Reliability. In: Transportation Network Analysis, eds. M.G.H. Bell and Y Iida, pp.179-192, John Wiley & Sons, Chichester.

10. Nicholson, A.; Schmöcker, J.-D.; Bell, M.G.H. and Iida, Y (2003) Assessing Transport Reliability: Malevolence and User Knowledge. In: The Network Reliability of Transport, eds. M.G.H. Bell and Y. Iida, Proceedings of the 1st International Symposium on Transportation Network Reliability, pp.1-22. Elsevier Science, Kidlington.

11. Dalziell, E.; Nicholson, A.J. (2001) Risk and impact of natural hazards on a road network. Journal of Transportation Engineering, 127(2), pp. 159-166.

12. LLeras-Echeverri, G. and Sanchez-Silva, M. (2001) Vulnerability analysis of highway networks, methodology and case study. Transport 174 (4), pp.223-230.

13. Srinivasan, K, (2002) Transportation Network Vulnerability Assessment: A quantitative framework. Transportation Security Papers 2002. White paper, Vanderbilt University/ Southeastern Transportation Center.

14. Husdal, J (2004) Reliability/vulnerability versus Costs/Benefits. Proceedings of the 2nd International Symposium on Transportation Network Reliability, Queenstown and Christchurch, New Zealand, 20-24 August 2004, p. 180-186.

15. Cova, T.J., and Conger, S. (2004) Transportation hazards, In: Handbook of Transportation Engineering, ed. M. Kutz, pp. 17.1-17.24, McGraw Hill, New York

16. Adler, H.A. (1987) Economic Appraisal of Transport Projects. Johns Hopkins University Press.

17. Nas, T. (1996) Cost-Benefit Analysis, Theory and Applications. Sage Publications

18. McHarg, I (1967) Where should Highways go? Landscape Architecture, 57, pp. 179-181

19. Bråthen, S and Lægran, S (2004) Bottlenecks in Road Freight Transport in Norway (Norwegian). Project report no. 245701,. Sweco Grøner, Oslo, Norway

20. Aven, T and Kørte, J. (2003) On the use of risk and decision analysis to support decision-making. Reliability Engineering and System Safety, 79, pp. 289-299

REFERENCE

Husdal, J. (2005) The vulnerability of road networks in a cost-benefit perspective. Paper presented at TRB2005, the Transportation Research Board Annual Meeting 2005, Washington DC, USA, 9-13 January 2005.

READ ONLINE

Jan Husdal TRB 2005

RELATED

• husdal.com:
  The reliability and vulnerability of transportation lifelines
• husdal.com:
  Reliability and vulnerability – a non-issue in cost-benefit analyses?
• husdal.com:
  Reliability and vulnerability versus costs and benefits (INSTR2004)
• husdal.com:
  Reliability and vulnerability versus costs and benefits (ETC2004)
• husdal.com:
  Transport network vulnerability – which metrics should we use? (NECTAR2006)
• husdal.com:
  Are roads more important than computers?