As of October 1, 2016, the SWUTC concluded its 28 years of operation and is no longer an active center of the Texas A&M Transportation Institute. The archived SWUTC website remains available here.

600451-00082

SWUTC Research Project Description

Real Time Optimization of Passenger Collection for Commuter Rail Systems (continuation of 600451-00068)

University: University of Texas at Austin

Principal Investigator:
Randy Machemehl
Department of Civil and Environmental Engineering
(512) 471-4541

Project Monitor:
Tod Hemingson
Vice President for Strategic Planning and Development
Capital Metropolitan Transit

Funding Source: USDOT and State of Texas General Revenue Funds

Total Project Cost: $155,725

Project Number: 600451-00082

Date Started: 5/1/12

Estimated Completion Date: 12/31/13

Project Summary

Project Abstract:
Urban commuter rail transit systems, by definition, are built upon existing rail right-of-way thereby enabling low cost rapid construction and implementation. Existing freight railroad tracks usually do not connect to ideal station locations within the central city area. Therefore, such systems require very effective collector-distributor systems to allow riders to conveniently reach their central city destinations. Previous studies on optimizing collector-distributer systems mostly offer optimal fixed route solutions based upon static demand estimates. Real time optimization using real time demand data eliminates the need for estimating service demand and accommodates urban environment dynamics more effectively. Although the research team has recently developed a real time optimization procedure for the one to many type passenger distribution problem, similar problem exists for the passenger collection activity that brings passengers back to the commuter rail station. Although the origin for all distribution trips is the transit station, origins for collection trips are scattered through the service area. Passengers who use the distribution bus service may expect to have service provided to the location where they alighted the bus, however, the optimally located routes and stops for distribution are likely not optimal for passengers being collected at any specific time. Although the optimization problem for collection is similar to the distribution problem, several important behavioral issues must be included and this clearly changes and complicates the problem. A solution method for the many to one passenger collection problem will be developed using an heuristic solution procedure to enable real time optimization of collector routes and stop patterns.

Project Objectives:
Heuristic algorithms will be developed for real time optimal routing and stop locations for the collection element of collector-distributor systems serving commuter rail transit stations. These algorithms will be evaluated and compared using a set of test problems. The procedure will be designed around real time passenger demand information acquired through a smart phone application that passengers can voluntarily use. Behavioral questions regarding passenger response to real time route and stop locations will be examined and findings will be used to guide algorithm development.

Task Descriptions:
Task 1. Background Literature Review
The scope of the preliminary scanning of relevant published work developed as part of this proposal will be extended to include problem characterization and solution concepts. This review will be documented to form the basis of the work of the subsequent tasks.

Task 2. Preparation of Prototype Network
A test bed or prototype network will be devised to serve as a primary development environment. The network development process will include a procedure for identifying the level of detail needed for algorithm testing. Possible passenger pick-up locations, or potential collector bus stop locations will be determined as part of this task. Average travel times will be assumed on all links according to their area type and functional classification.

Task 3. Determining the Service Area
Next, a procedure for estimating the “service area” that can be served reasonably by a feeder route will be developed. The size of the service area depends on route travel time constraints (links that cannot be covered in the framework of time allocated for the travel should be eliminated from the base network). Route travel time constraints are derived from traveler behavioral considerations of maximum total trip times as well as schedule constraints associated with the line haul commuter rail mode.

Task 4. Real Time Passenger Demand Information (Model Input)
Application of the product of this research will depend upon real time collection of passenger demand information through smart phone, basic cell phone or other communication means. For development purposes, these data will be initially assumed. However, a human behavioral issue must be considered in that all passengers may not wish to call for service. Some may prefer to be able to use a fixed stop location so the methodology may need to provide at least a minimal number of fixed stop locations to accommodate those who cannot or choose not to call for real time service. This task will evaluate the benefits and costs to the overall program of providing fixed stop locations in addition to demand responsive stop locations. Considering everyday advances in technology and growing numbers of people who use modern communication technologies, assuming that rail transit users can communicate their location information via their smartphones is plausible. However, other methods of communication such as calls, internet, and installation of information keypads at some designated locations will be explored.

Task 5. Model Formulation
This task involves formulating the problem in mathematical terms. It will mathematically show formulations of objective function, input parameters, constraints, and decision variables. This formulation will be the basis for developing the Tabu heuristic algorithm and other heuristic solution methods will also be explored.

Task 6. Developing Heuristic Algorithm
This task involves creating the GA, SA, and TS optimization search algorithms and encoding using Matlab, Java, or C++ languages. The problem can be characterized in a variety of different ways using any of the three heuristic procedures but particularly the GA implementation provides many options. The advantages and disadvantages of each method will be explored considering the likelihood of finding the global optimal answer and the processing times.

Task 7. Testing the Algorithm
After completion of encoding, several test problems, will be performed to evaluate the results of this analysis. This process will involve solution of a range of network problems that are small enough to allow finding the global optimum value through complete enumeration so that results of the heuristic can be compared to the true answer across a series of different networks.

Task 8. Presentation of Results
At least two technical papers will be developed, presented and published nationally to advise potential users of the new capability. The research team will work with the Capital Metropolitan Transit Authority in Austin where a commuter rail line has been recently implemented. A comprehensive research report will be developed to document all findings.


Implementation of Research Outcomes:
Commuter rail systems are being introduced into many urban areas as an alternative mode to automobiles for commuting trips. The shift from the auto mode to rail mode is anticipated to greatly help alleviate traffic congestion in urban road networks. However, the right-of-way of many existing commuter rail systems is usually not ideally located. Since the locations of rail systems were typically chosen long ago to serve the needs of freight customers, the majority of current commuter rail passengers have to take a non-walkable connecting trip to reach their final destinations after departing even the most conveniently located rail stations. To make rail a more viable, competitive commuting option, a bus feeder or circulator system is proposed for seamlessly transporting passengers from their departing rail stations to final work destinations. The primary research challenge in modeling such a bus circulator system is to optimally determine a bus route and stop sequence for each circulating tour using the real-time demand information. In this research, we termed this joint routing and stop optimization problem the circulator service network design problem, the objective of which is to minimize the total tour cost incurred by bus passengers and operators while minimizing the walk time of each individual bus passenger.

From the application perspective, the technology developed by this research provides a new tool which permits real-time optimization of real-time starts and stops.  From a theoretical perspective, the solution method is a breakthrough in terms of speed for solving optimization problems.  The study results have been discussed with Austin’s Capital Area Metropolitan Transportation Agency.

Products developed by this research:

Model and Method Developed:  A bi-level nonlinear mixed integer programming model was constructed and a tabu search method with different local search strategies and neighborhood evaluation methods was then developed to tackle the circulator service network design problem.

Presentation:  Real-Time Optimization of Passenger Collection for Commuter Rail Systems, Yao Yu and Randy Machemehl, presented at and published in the Proceedings of the Canadian Society of Civil Engineers 10th International Specialty Conference on Transportation, May 2014.

Publication Pending:  Demand-responsive Transit Circulator Service Network Design, Yao Yu, Randy B. Machemehl, and Chi Xie, submitted for publication in Transportation Research Park E:  Logistics and Transportation Review (pending)

Journal Article in Preparation:  Optimizing Transit Circulator Service Networks for Commuter Rails, Yao Yu, Randy B. Machemehl, and Chi Xie, to be submitted to the Journal of American Society of Civil Engineers.

Impacts/Benefits of Implementation:
This project will be helpful for public transportation agencies that seek supporting materials for developing plans for new technology adoption and implementation.  The results are also applicable to anyone dealing with solution methods for optimization problems.  This multidisciplinary community would include mechanical engineers, electrical engineers and business managers.

The findings in this report could serve to improve public knowledge and attitudes, as well as changing behavior, practices, decision-making and/or policies towards emerging transportation technologies.

Web Links:
Final Technical Report