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.


SWUTC Research Project Description

Zone/Fleet Sizing for MAST (Mobility Allowance Shuttle Transit) Services

University: Texas A&M University

Principal Investigator:
Luca Quadrifoglio
Texas Transportation Institute
(979) 458-4171

Project Monitor:
Dr. Maged Dessouky
Daniel J. Epstein Department of Industrial and Systems Engineering
University of Southern California

Funding Source: USDOT

Total Project Cost: $65,925

Project Number: 600451-00026

Date Started: 1/7/13

Estimated Completion Date: 12/31/14

Project Summary

Project Abstract:
This project investigates critical design characteristics of the Mobility Allowance Shuttle Transit (MAST) service or “route-deviation”.  The service is more flexible than fixed route transit and more cost efficient than demand responsive transit.  Vehicles follow a base fixed route, with one or more mandatory checkpoints, but are allowed to deviate from their path to serve passengers at their desired locations.  Vehicles are allowed some slack time for deviations within a predetermined service area.  In particular, the research team will develop analytical tools to identify the proper size/shape of the zone served by each line and the fleet size assigned to each line to optimize a properly defined objective function.  Results will be a preliminary step towards implementation and testing in large-scale applications.  The ultimate goal would be enhancing mobility and livability of communities.

Project Objectives:
Only the simple single-vehicle/single-line/single-zone case has been studied so far and the investigation of a multi-vehicle/line/zone large-scale network implementation of MAST has yet to be addressed by researchers and practitioners. The general problem of designing a transit system can roughly be framed as having three main variables: its demand, its operating costs and its service quality.  It would be desirable to simultaneously minimize the operating costs and maximize the service quality.  But, in general, these objectives are competing.  The general trend of the cause/effect relationships between them may be intuitive; however, an analytical quantification of it is not trivial and is necessary to allow decision makers to perform wise tradeoffs between the design variables.

We are proposing to develop general relationships for two of the main decision variables for design purposes, in order to respond to this currently unanswered questions:

  • How many zones and of what size/shape is recommendable for maximizing the overall performance of a MAST system operating in a service area with a given demand distribution?
  • How many and what size of vehicles are recommendable within each zone for the same purpose?

The research will be performed by developing appropriate analytical relationships verified/validated by a set of simulation analyses.  The PI has previous experience and publications in using such methodology, similarly applied to other systems.  Previous PI’s research on MAST will also be built upon for conducting this study.

While we cannot reasonably expect precise and unique numerical answers, we can confidently predict that the expected outcome of this research will be a set of recommendations for different plausible scenarios and guidelines on how to properly make the described critical decisions.

Task Descriptions:
Task 1.  Literature Review

The study will collect as much information as possible about the topic and related research and/or applications to integrate and update the already performed literature review.

Task 2: Develop the analytical tool to identify fleet size
To optimize the resources for operating these systems, we will develop an analytical model to identify the critical demand level (and related time of the day) at which it would be recommendable to dispatch additional vehicles. Continuous approximations and logical analytics will be used for performing this task.  The aim is to maximize the overall performance of the system, a balance between operating costs and service level.  Decision makers will be able to choose the values of the weights assigned to each one of these two terms in order to select the appropriate demand level.

Task 3.  Develop the analytical tool to identify optimal zone design
The Multi MAST service is a very complex system to be scheduled, as it is partially fixed and partially on-demand. All is complicated by the fact that more than one vehicle is operating within the service area. In designing such systems for large communities, planners may divide the whole service area into zones for easier management of the operation, to reduce operating cost, and to provide a better level of service to customers. In each zone, an independent MAST line would provide the service to its customers. For example, the best number of zones is difficult to determine because the balance between operating costs and service quality is frequently difficult to evaluate, especially within areas with low and sparse demands. However, a non-optimal structure might be adopted for lack of zone design, simply because these services are considered a niche market. However, trends suggest that these services will progressively increase their market share and importance within transit agencies, demanding a more rigorous and methodological design approach to the problem.

This research study builds on previous work performed by the PI, in which an analytical model was developed to determine the best operating policy for adoption in a residential zone to maximize the level of service. This project develops an analytical model to help planners determine the optimal number of zones while balancing customer level of service and operating costs. These are powerful tools that can help solve the complex MAST design problem.

Task4.  Construct a Simulation Model
In this stage, we will develop a simulation model to analyze the performance of the developed analytical tools in Task 2 and Task 3. Simulated demand will initially be generated to evaluate simple scenarios. Several scenarios will be considered to perform sensitivity analyses and evaluate alternative options.  Real case scenarios with real demand data will also be considered.

Task 5.  Deliverables
The deliverables of this project will be the final report that will be provided to the South West University Transportation Center.

Implementation of Research Outcomes:
The mobility allowance shuttle transit (MAST) system is an innovative concept that allows transit vehicles to deviate from a fixed route consisting of a few mandatory checkpoints to serve on-demand customers within a predetermined service area, and thus can be both affordable and convenient enough to attract the general public. Through this effort, researchers developed analytical results for the waiting time probability distribution and its expected value as well as the expected ride time for different types of customers in terms of the system parameters for both 1-MAST system and multi-vehicle MAST (m-MAST).

The results of this project provides the basis for evaluating the performance of MULTI-MAST transit systems, to better plan for resource allocation (such as fleet sizing) and deployment arrangements (such as zoning).

Impacts/Benefits of Implementation:
Results produced by this project have the potential to impact and improve current practice in decision making in the transportation industry, specifically for planning and design of transit systems. This could lead to cost saving and performance improvement of such system, potentially improving livability and quality of life.

More in depth and accurate assumptions in assuming random customer demand provides for better planning.  This can be used in other systems dealing with similar random demand, such as in the service industry other than transit.

Web Links:
Final Technical Report