Army Corps of Engineers Waterway Projects.
Posted: Mon Apr 01, 2002 1:45 pm
Chariya,
I would suggest reading the full National Academy of
Science Report regarding the allegations surrounding
the "Inland Navigation System Planning: The Upper
Mississippi River-Illinois Waterway" project
(UMR-IWW).
http://www.nap.edu/books/0309074053/html/
For the larger community, Chapter 5 of the report,
which outlines "the current state of the art in
modeling the uncertainty associated with forecasting
future events" they never mention System Dynamics.
In fact, I searched the entire book for even one
mention of using System Dynamics and found nothing.
What gives? Surely, SD has something to offer to this
type of complex analysis?
For Chariya and others interested in the details . . .
The report starting at page 30 begins discussing the
role of the environment as being an important
consideration in planning to expand locks (WOW -- what
an insight, to think that the environment might be
impacted by expanding locks!) As the report puts it:
"A true systems view of the UMR-IWW would acknowledge
that a thourough understanding of the current state of
the Upper Mississippi Rive and floodplain ecosystem
must be related to structural modifications in the
Upper Mississippi River basin, to operations of the
UMR-IWW navigation system, and to ecosystem responses
to a range of natural and anthropogenic influences
within the Upper Mississippi River basin. The
feasibility studey would also benefit from a
clarification of the roles of the environmental
studies in the decision-making process regarding lock
extensions."
The report talks about the two models developed: (1) a
spatial equilibrium model and (2) the ESSENCE model.
What is spatial equilibrium modeling (SEM)? Sounded
strange and mysterious, so I did some quick reading.
SEM is basically an economic approach to trying to
understand complexity. It is evidently often used to
analyze international agricultural trade. Based on
what I found, spatial equilibrium models usually
operate under assumptions of perfect competition, but
from a quick look on google, its obvious that people
are adapting it to deal with imperfect competition.
The equation you mention is related to the simulation
ESSENCE model (an excel spreadsheet model containing
two main components and one smaller component that
calculated the economic benefit of the waterway).
According to the report, these components are then
"reconciled using Excels "solver" function (page 36
of the report).
Q=[(R-y)/(R-W(o))]^N * Q(o)
To summarize the reports comments on the four key
weaknesses of the Corps Model:
1. Forecasting of domestic and foreign commodity
production and use is based mostly on extrapolations
of historical data (The Corp evidently used USDA
forecasts). Forecasts made in 1994 for 1995-2000 were
inaccurate (example: model forecasted increases in
grain exports: reality was that total exports of corn
and wheat trended downward). From what I read, I
believe that the orginal Corps report was seeking to
create projections for *50* years into the future.
There was some confusion regarding which data to use
for regression analysis. As I read this report, I
shuddered to think about using Excel to create this
type of model.
****Does anyone know if the USDA is now using SD
modeling software to create large-scale agricultural
sector models?
2. Determining the sensitivity of waterway tonnage to
barge tarifffs was based on assumptions that were not
realistic. In the equation, the exponent N represented
an empiraclly determined constant that was open to
serious debate (to put it mildly).
3. Calculating National Economic Development benefits.
This is a whole section that calls into question data
used in the model and certain techniques. Enjoy
reading!
4. Modeling lock congestion was done using a simple
queing model to model increase in delays at locks in
response to traffic increases. They recommend that
simulation modeling (rather than queing theory) be
used to analyze lock performance. Do they mean SD
simulation modeling or are they talking about
something else? I dont know. Does anyone?
Best,
Justin
From: Justin Lyon <justin1028@yahoo.com>
Other interesting tidbits . . .
http://www2.mvr.usace.army.mil/umr-iwws ... background
http://www2.mvr.usace.army.mil/umr-iwws ... n=home.faq
I would suggest reading the full National Academy of
Science Report regarding the allegations surrounding
the "Inland Navigation System Planning: The Upper
Mississippi River-Illinois Waterway" project
(UMR-IWW).
http://www.nap.edu/books/0309074053/html/
For the larger community, Chapter 5 of the report,
which outlines "the current state of the art in
modeling the uncertainty associated with forecasting
future events" they never mention System Dynamics.
In fact, I searched the entire book for even one
mention of using System Dynamics and found nothing.
What gives? Surely, SD has something to offer to this
type of complex analysis?
For Chariya and others interested in the details . . .
The report starting at page 30 begins discussing the
role of the environment as being an important
consideration in planning to expand locks (WOW -- what
an insight, to think that the environment might be
impacted by expanding locks!) As the report puts it:
"A true systems view of the UMR-IWW would acknowledge
that a thourough understanding of the current state of
the Upper Mississippi Rive and floodplain ecosystem
must be related to structural modifications in the
Upper Mississippi River basin, to operations of the
UMR-IWW navigation system, and to ecosystem responses
to a range of natural and anthropogenic influences
within the Upper Mississippi River basin. The
feasibility studey would also benefit from a
clarification of the roles of the environmental
studies in the decision-making process regarding lock
extensions."
The report talks about the two models developed: (1) a
spatial equilibrium model and (2) the ESSENCE model.
What is spatial equilibrium modeling (SEM)? Sounded
strange and mysterious, so I did some quick reading.
SEM is basically an economic approach to trying to
understand complexity. It is evidently often used to
analyze international agricultural trade. Based on
what I found, spatial equilibrium models usually
operate under assumptions of perfect competition, but
from a quick look on google, its obvious that people
are adapting it to deal with imperfect competition.
The equation you mention is related to the simulation
ESSENCE model (an excel spreadsheet model containing
two main components and one smaller component that
calculated the economic benefit of the waterway).
According to the report, these components are then
"reconciled using Excels "solver" function (page 36
of the report).
Q=[(R-y)/(R-W(o))]^N * Q(o)
To summarize the reports comments on the four key
weaknesses of the Corps Model:
1. Forecasting of domestic and foreign commodity
production and use is based mostly on extrapolations
of historical data (The Corp evidently used USDA
forecasts). Forecasts made in 1994 for 1995-2000 were
inaccurate (example: model forecasted increases in
grain exports: reality was that total exports of corn
and wheat trended downward). From what I read, I
believe that the orginal Corps report was seeking to
create projections for *50* years into the future.
There was some confusion regarding which data to use
for regression analysis. As I read this report, I
shuddered to think about using Excel to create this
type of model.
****Does anyone know if the USDA is now using SD
modeling software to create large-scale agricultural
sector models?
2. Determining the sensitivity of waterway tonnage to
barge tarifffs was based on assumptions that were not
realistic. In the equation, the exponent N represented
an empiraclly determined constant that was open to
serious debate (to put it mildly).
3. Calculating National Economic Development benefits.
This is a whole section that calls into question data
used in the model and certain techniques. Enjoy
reading!
4. Modeling lock congestion was done using a simple
queing model to model increase in delays at locks in
response to traffic increases. They recommend that
simulation modeling (rather than queing theory) be
used to analyze lock performance. Do they mean SD
simulation modeling or are they talking about
something else? I dont know. Does anyone?
Best,
Justin
From: Justin Lyon <justin1028@yahoo.com>
Other interesting tidbits . . .
http://www2.mvr.usace.army.mil/umr-iwws ... background
http://www2.mvr.usace.army.mil/umr-iwws ... n=home.faq
