Geoscience Reference
In-Depth Information
In answering those questions, scientific reasoning and methods developed in
game theory and operations research could shed light on many issues. Many
megacity issues can be formulated by using the theoretical framework of
Stackelberg's leader-follower game where such organization as the metropolitan
planning organization (MPO) is the leader and each city is a follower. Many such
issues could also be formulated in multi-level programming problems (e.g. see
Ben-Ayed et al.
1992
; Suh and Kim
1992
).
19
The most effective transportation plans at the megacity level are not necessarily
the sum of the most effective transportation plans for the individual cities. For
example, the total minimum travel cost of a system as a whole in a megacity is not
the sum of the minimum travel costs of the individual cities. Knowing that each
individual city's transportation plans are usually based on findings regarding a
traveler's perceived minimum cost travel patterns, an example of a multi-level
programming model could be:
• Upper Level Objective Function: Minimize Total System's Travel Time
• Lower Level Objective Function: Minimize Individual City's Total Travel Time
- Subject To: Minimize Each Traveler's Perceived Travel Time
Under this formulation, the system's optimum can be found assuming that
MPOs and/or COGs function as the leader while preserving each individual city's
functions as a follower, but maximizing each individual city's welfare.
17.5.3 Investment Decision Making Processes for Transport
Infrastructure
Currently, making decisions about investing in transport infrastructure are based on
cost-benefit analysis results by comparing the anticipated benefits and costs. The
costs usually include land acquisition, construction, and maintenance cost over the
life span of the infrastructure. The benefits are more difficult, but normally include
savings in travel time and maintenance costs, accident reductions and environmen-
tal benefits. Since the benefit is assumed to occur to each individual traveler, the
more travelers use the infrastructure, the more benefits.
Often, a number of alternative scenarios are evaluated by estimating the net
benefit and the scenario that results in the highest net benefit has been chosen for
implementation. If there is only one alternative scenario, the net benefits are
estimated for each case with and without the scenario, providing the decision
maker clues for go or no-go for the project. Frequently, transportation demand
models are used for the estimating, forecasting number of passenger-car unit
equivalent unit (PCU) for each link in the study area for the cost-benefit analysis.
In the past, planners and decision makers focused on the elimination of
congestion as the main goal for solving transportation problems by expanding
19
The entire Volume 34 of the Journal of Operations Research, 1992 includes articles on the
development and use of multi-level programming issues.
