Civil Engineering Working Papers

Trading public transport travel demand for electronic coupons through mobile device fare collection

Joseph YJ Chow — Mon, 18 Mar 2013 05:55:09 PDT

Mobile technologies are generating new business models for urban transport systems, as is evident from recent startups cropping up from the private sector. Public transport systems can make more use of mobile technologies than just for measuring system performance, improving boarding times, or for analyzing travel patterns. Unlike earlier studies on mobility credits, a new transaction model is proposed for public transport systems where travelers are allowed to pre-book their fares and trade that demand information to private firms. In this public-private partnership model, fare revenue management is outsourced to third party private firms such as big box retail or large planned events (such as sports stadiums and theme parks), who can issue electronic coupons to travelers to subsidize their fares. This e-coupon pricing model is analyzed using marginal cost theory and shown to be quite effective for monopolistic firm participation, particularly for demand responsive transit systems that feature high cost fares, non-commute travel purposes, and a closed access system with existing pre-booking requirements. However, oligopolistic scenarios analyzed using game theory and network economics suggest that public transport agencies need to take extreme care in planning and implementing such a policy. Otherwise, they risk pushing an equivalent tax on private firms or disrupting the urban economy and real estate values.

Activity-based travel scenario analysis with routing problem reoptimization

Joseph YJ Chow — Mon, 01 Oct 2012 07:45:09 PDT

Activity-based travel scenario analysis and network design using a household activity pattern problem (HAPP) can face significant computational cost and inefficiency. One solution approach, called reoptimization, makes use of an optimal solution of a prior problem instance to find a new solution faster and more accurately. Although the method is generally NP-hard as well, the approximation bound is found to be tighter than a full optimization for several traveling salesman problem variations. To date, however, there have not been any computational studies conducted with the method, nor has there been any meta-heuristics designed with reoptimization in mind, particularly for generalized vehicle routing problems. A generalized, selective household activity routing problem (G-SHARP) is presented as an extension of the HAPP model to include both destination and schedule choice for the purpose of testing reoptimization. Two reoptimization algorithms are proposed: a simple swap heuristic and a new class of evolutionary algorithms designed for reoptimization, dubbed a Genetic Algorithm with Mitochondrial Eve (GAME). The two algorithms are tested against a standard genetic algorithm in a computational experiment involving 100 zones that include 400 potential activities (resulting in a total of 802 nodes per single-traveler household). Five hundred households are synthesized and computationally tested with a base scenario, a scenario where an office land use in one zone is de-zoned, and a scenario where a freeway is added onto the physical network. The results demonstrate the effectiveness of reoptimization heuristics, particularly GAME, and the capability of G-SHARP to capture re-allocations of activities and schedules with respect to spatiotemporal changes.

Nonlinear inverse optimization and link-based freight assignment of commodity and cyclic vehicle flows

Joseph YJ Chow et al. — Mon, 01 Oct 2012 06:30:14 PDT

An inverse nonlinear optimization approach is developed to solve a new parameter estimation problem class where demand and output flows are given or observed in a congested network. Unlike earlier work in the inverse optimization area, the problem is nonlinear and formulated with Karush-Kuhn-Tucker conditions. It is shown to be solvable using classical nonlinear optimization methods. This approach is used to systematically calibrate a new link-based variant of the STAN model which assigns commodity flows to cyclic vehicles, where freight facility parameters are typically difficult to observe. Commodities flow from origin to destination but vehicle path information is given up in favor of tracking cyclic patterns for endogenous empty hauls. The models are first tested on a small network with up to 54 transshipment activities. The assignment model is shown to be sensitive to supply side changes on links and transshipment facilities or to fuel cost changes, while the inverse model is tested for parameter recovery and similarity in scenario sensitivity. The calibration method is then applied to California with 1,058 transfer links with 2007 data, under the circumstance where empty haul patterns and some truck patterns are unavailable. The assignment model is validated with 2010 data. The appendix includes a first application of the inverse nonlinear programming method to the inverse traffic assignment problem to demonstrate its value in other potential fields.

A surrogate-based multiobjective metaheuristic and network degradation simulation model for robust toll pricing

Joseph YJ Chow et al. — Wed, 18 Jul 2012 10:13:33 PDT

From a problem-centric perspective, robust network design problems are systems engineering methods that share common research gaps. First, problem sizes are constrained due to the use of multi-objective solution algorithms that are notoriously inefficient due to costly function evaluations. Second, link disruptions at a network level are not easy to model realistically. In this paper a stochastic search metaheuristic based on radial basis functions is proposed for constrained multiobjective problems. It is proven to converge, and compared with conventional metaheuristics for four test problems. A scenario simulation method based on multivariate Bernoulli random variables that accounts for correlations between link failures is proposed to sample scenarios for a mean-variance toll pricing problem. Four tests are conducted on the classical Sioux Falls network to gain some insights into the algorithm, the simulation model, and to the robust toll pricing problem. The first test empirically measures the efficiency of the simulation algorithm and approximate Pareto set by obtaining a standard error in the -indicator measure for a given number of scenarios and iterations. The second test compares the dominance of the proposed heuristic’s solutions with a conventional multiobjective genetic algorithm by comparing the average epsilon-indicator. The third test quantifies the gap due to falsely assuming that link failures are independent of each other when they are not. The last test quantifies the value of having the flexibility to adapt a Pareto set of toll pricing solutions to changing probability regimes such as peak and off-peak hurricane or snow seasons.

On observable chaotic maps for queueing analysis

Joseph YJ Chow — Thu, 12 Jul 2012 15:06:18 PDT

A queueing model based on chaotic mapping offers a number of distinct advantages over both stochastic and static deterministic models. Depending on the type of chaotic map used, such a queue can capture transient behavior, intermittency, steady state behavior, and complex distributions in arrival rates. These characteristics are especially desirable in many queueing applications in transportation. Earlier studies resulted in chaotic queueing models that cannot be estimated using observed arrivals. An alternative queueing model is presented along with methods to specify the model, interpret its results, and estimate its parameters. The proposed queueing model uses chaotic maps of inter-arrival times to generate arrivals so that parameters can be calibrated with observable data. A sample queue based on the ergodic logistic map is presented. To calibrate the mapping based on observed data, a joint parameter and state estimation algorithm is presented using the method of successive averages. An illustration is made with two connected queues to show how a purely deterministic queueing network can still result in a joint invariant distribution. The results offer a positive view of this method and its applicability to queueing problems, particularly in the field of transportation and dynamic network loading.