Paired-Line Hybrid Transit

Paired-Line Hybrid Transit was the first in a series of “hybrid-transit” studies conducted by my group using a stylized design model.  This line of work, funded by an National Science Foundation between   2013 and 2016, was initiated by Peng Chen in his PhD thesis.  The main idea is to pair a demand-adaptive service with a fixed-route service so that the transit system can leverage the advantages of both while avoiding their drawbacks.  The paper was published in Transportation Research Part B in 2017.

For preprint, check  Hybrid Transit System Design_Journal_2.0


Abstract: This paper proposes and analyzes a new transit system that integrates the traditional fixed-route service with a demand-adaptive service. The demand-adaptive service connects passengers from their origin/destination to the fixed-route service in order to improve accessability. The proposed hybrid design is unique in that it operates the demand-adaptive service with a stable headway to cover all stops along a paired fixed-route line. Pairing demand-adaptive vehicles with a fixed-route line simplifies the complexity of on-demand routing, because the vehicles can follow a more predictable path and can be dispatched on intervals coordinated with the fixed-route line. The design of the two services are closely
coupled to minimize the total system cost, which incudes both the transit agency’s operating cost and the user cost. The optimal design model is formulated as a mixed integer program and solved using
a commercially available metaheuristic. Numerical experiments are conducted to compare the demand adaptive paired-line hybrid transit (DAPL-HT) system with two related transit systems that may be considered its special cases: a fixed-route system and a flexible-route system. We show that the DAPL-HT system outperforms the other two systems under a wide range of demand levels and in various scenarios of input parameters. A discrete-event simulation model is also developed and applied to confirm the correctness of the analytical results.

A theory of justice

Until I come across John Rawls’ book, I have never thought the word Theory can be associated with Justice (公正).  I was reluctant to commit my leisure time to reading “theories” (my job gives me plenty already), but my better judgement was overridden by the charisma of the loaded buzzword of our time.  Don’t get me wrong: the book is worth reading.  However, getting through five hundred pages of hard (and dry) reasoning and argument—with few stories or quips to catch a break—was quite a mental exercise.  Well, here is what I have learned.

A theory of justice is a set of principles designed to resolve the conflict of interests arising from human cooperation.  These principles form the basis for regulating social behaviors and arranging political institutions.   Rawls’ theory, which earned him the reputation as one of the most influential political philosophers of the 20th century, is dubbed “Justice as Fairness”.   It consists of three principles. The first principle (A) guarantees an equal right to “most extensive” basic liberties.  The other two state social and economic inequalities are just if and only if they (B) are attached to positional goods open to all under conditions of equal opportunity, and (C) maximize primary goods enjoyed by the least advantaged members of society.  The three principles follow a lexical order: the equal liberty principle (A) takes priority, followed by the equal opportunity principle (B) and the difference principle (C).

At first glance justice as fairness seems decisively egalitarian.  Rawls advocates compensation for the “undeserved inequalities” from birth and natural endowment.  Under the difference principle (C), “society must give more attention to those with fewer native assets and to those born into the less favorable social positions”.  An example is spending greater resources on “the education of the less rather than the more intelligent”.  In fact, Rawls has gone so far as to assert the willingness to make an effort is also contingent upon social circumstances. In other words, even inequalities caused by laziness may be undeserved and thus warrant social redress.    To be sue, Rawls does not support eradicating inequalities all together. Instead, inequalities are to be tolerated in so far as they benefit everyone, especially the most disadvantaged.    Moreover, positional goods should be distributed based on the equal opportunity principle, and if doing so leads to inequalities, so be it.  Thus, Rawls seems to prefer meritocracy to equal-outcome when it comes to positional goods. There is a catch though: everyone must “have reasonable opportunity to acquire the skills on the basis of which merit is assessed”.  Needless to say, compensations are necessary to meet such a requirment.

Rawls contends justice as fairness is preferred by moral persons at the original position.   He assumes moral persons know what is right (i.e., have a sense of justice) and what constitutes their own good.  As such, moral persons are entitled to equal justice. As Rawls puts it, “those who can give justice are owed justice”.   The original position, on the other hand, wraps everyone with a veil of ignorance. From behind this veil, nobody knows their social status or natural asset.  As a result, they cannot and will not tailor the principles to their own advantage.   Rawls’ moral persons are not altruistic, but mutually disinterested and rational.  This means they (i) strive to maximize their own good, as defined by the rational plan of life, and (ii) are indifferent to the good of others, in both absolute and relative terms.    Rawls’ faith in the better angels of our nature is admirable, but I am not sure our cravings for justice can resist the incessant onslaught of envy, vanity, and self-serving biases.   A society built on justice as fairness may be ideal, even optimal. But will it be stable?  Perhaps that’s what Benjamin Franklin had in mind when he proclaimed, “A republic if YOU CAN KEEP IT!”

Rawls staunchly opposes utilitarianism, the idea that a society is property arranged when its institutions maximize the net balance of welfare.  He believes “each person possesses an inviolability founded on justice that even the welfare of society as a whole cannot override.”   Therefore, justice “denies that the loss of freedom for some is made right by a greater good shared by others.”   These words ring so true as I am, like many of my friends are, trying to understand what the heck is going on in Shanghai.

Mitigating TNC-induced traffic congestion

While the e-hail service offered by TNCs is widely credited for boosting productivity and enhancing level of service, its adverse traffic impact in already-congested city centers has drawn increased scrutiny.   Several cities have started to implement  policies aiming to mitigate the traffic impact induced by excessive TNC operations.   The purpose of this study is to support such policy analysis by developing a model that captures the complex interactions among various stakeholders (riders, drivers and the platform) and those between them and the regulator.   Please read the abstract below for main findings.

The paper was recently published in Transportation Research Part A.   A preprint can be downloaded here.


Abstract: This paper analyzes and evaluates several policies aiming to mitigate the congestion effect a Transportation Network Company (TNC) brings to bear on an idealized city that contains a dense central core surrounded by a larger periphery. The TNC offers both solo and pooling e-hail services to the users of public transport. We develop a spatial market equilibrium model over two building blocks: an aggregate congestion model describing the traffic impact of TNC operations on all travelers in the city, including private motorists, and a matching model estimating the TNC’s level of service based on the interactions between riders and TNC drivers. Based on the equilibrium model, we formulate and propose solution algorithms to the optimal pricing problem, in which the TNC seeks to optimize its profit or social welfare subject to the extra costs and/or constraints imposed by the congestion mitigation policies. Three congestion mitigation policies are implemented in this study: (i) a trip-based policy that charges a congestion fee on each solo trip starting or ending in the city center; (ii) a cordon-based policy that charges TNC vehicles entering the city center with zero or one passenger; and (iii) a cruising cap policy that requires the TNC to maintain the fleet utilization ratio in the city center above a threshold. Based on a case study of Chicago, we find TNC operations may have a significant congestion effect. Failing to anticipate this effect in the pricing problem leads to sub-optimal decisions that worsen traffic congestion and hurt the TNC’s profitability. Of the three policies, the trip-based policy delivers the best performance. It reduces traffic congestion modestly, keeps the TNC’s level of service almost intact, and improves overall social welfare substantially. The cruising cap policy benefits private motorists, thanks to the extra congestion relief it brings about. However, because other stakeholders together suffer a much greater loss, its net impact on social welfare is negative. Paradoxically, the policy could worsen the very traffic conditions in the city center that it is designed to improve.