# Transportation Deployment Casebook/About

This Casebook presents the results of an assignment for CE5212/PA5232 Transportation Policy, Planning, and Deployment to describe the lifecycle of a transportation technology or mode.

## Contents

## The Assignment[edit]

Recall that the cycle of technology includes a birthing phase, a growth-development phase, and a mature phase (and perhaps a declining phase). The stage of the life-cycle, it has been argued, determines the nature of transportation policy-making -- both the problems faced and the responses to these problems. In this assignment, you are to research and reflect upon the life-cycle of a transportation mode. Your final product should be about 15 pages of single-spaced 12 point Times New Roman text, including tables and charts.

Your initial step is to select a mode (or transportation or related technology). As long as the technology you pick can be related to the movement of people, goods, or ideas it is fine. If you have questions, contact your instructor.

Then the assignment has two parts: Quantitative and Qualitative. Each part is worth 50% of the assignment.

### Quantitative[edit]

Overview the life-cycle of the mode. Using S-curves (status vs. time), identify the periods of birthing growth, and maturity. For status use variables that reflect the level of deployment or use of the mode (number of vehicles, kilometers of track, passenger-kilometers traveled). You may develop these curves for the US, some other country, or geographical unit, as data availability and your focus dictate. Use the data to estimate a three-parameter logistic function:

where:

- is the status measure, (e.g. Passenger-km traveled)
- is time (usually in years),
- is the inflection time (year in which 1/2 K is achieved),
- is saturation status level,
- is a coefficient.

and are to be estimated.

Graph the model and the data. How accurate is the model?

Interpret your results, and use them to help affix dates to the birthing, growth, and maturity stages of the life-cycle.

Sources of data vary. There is sufficient data on the world wide web to do this assignment for many technologies and modes, though there are obviously modes and technologies for which this will be difficult. For a good starter source, try the Bureau of Transportation Statistics ( http://www.bts.gov ). However, make sure that you get data that goes to the birthing phase of the technology, many of their data series only go back to 1960. For other data sources, the world wide web and the university library, as well as the Mn/DOT library are good places to go.

#### Get your data[edit]

GET YOUR DATA FIRST.

#### How to estimate a model:[edit]

An example spreadsheet is at:

http://nexus.umn.edu/Courses/ce5212/RailReinvented.xls

Basically, this is an exercise in curve fitting. There are better and worse ways to do curve fitting, one (a better way, but not the best way) is shown in the example spreadsheet.

Worse ways to do curve fitting include random trial and error (pick some values, see how close the curve is, and adjust the values).

Better ways involve using a formal statistical procedure (e.g. Ordinary Least Squares Regression). Fortunately, regression is a routine procedure found in many statistical packages, as well as in spreadsheets. To do a regression in an excel spreadsheet, you can find the tool under the TOOLS menu in the DATA ANALYSIS option. (If DATA ANALYSIS does not appear on your menu, go to TOOLS/ADD INS, and Check the DATA ANALYSIS Toolbox).

A single variable linear regression simply estimates the coefficients c and b in a model of the form:

The question is then, what is y (your dependent variable) and what is x (your independent variable)

In the example spreadsheet

The Derivation is given on the spreadsheet.

The number of passengers is known from the data, as is the year. (LN is the natural log function). K however is unknown in this case (High Speed Rail is not yet fully mature, traffic volumes are not saturated). Thus a number of regressions, with various trials for K were done.

The goodness of fit of the model is explained by the R-squared and the t-statistic. You want an R-squared close to 1.0, but are unlikely to get an exact value. You want t-statistics on your estimated variables to be as high as possible, and generally higher than 2 (which indicates the variable is statistically significant at the 95% confidence level).

The constant term, along with the coefficient on the independent variable give you your t0.

### Qualitative[edit]

- Describe the mode. What are its essential technological characteristics, its main advantages, and its main markets.
- Set the scene prior to the advent of the mode. What other modes were available? What were their limitations? How were markets for transportation evolving? How did these factors stir interest in new possibilities?
- Describe the invention of the mode / technology. What different types of technological expertise were brought together? How? How was the shift from the initial design altered in the face of early experience? Describe the shift from the initial technology to the predominant technology. Remember that technology refers both to hardware (physical artifacts) and software (the way the artifacts are used to produce transportation).
- Describe early market development. What were the initial market niches? What roles did functional enhancement (serving existing markets better) and functional discovery (serving new markets) play in market development?
- Assess the role of policy in the birthing phase. Describe how policies from precursor models were borrowed, and how other policies were innovated. Identify policies that were embedded and policies that were imposed or sanctioned by government. Identify policies that were "locked in" during this time.
- Describe the growth of the mode. What roles did the public and private sectors play in the growth? What policy issues arose, and how were they resolved? How did the policy environment influence policymaking in this period.
- Describe development during the mature phase of the mode. Describe attempts to adapt the mode changing markets, competitive conditions, and policy values. Describe how "lock-in" has constrained these adaptations. Identify any opportunities you see to "re-invent" the mode so that it can better serve the needs of today and tomorrow.

### Some Useful Online Datasets[edit]

- Statistical Abstracts
- Bureau of Transportation Statistics
- National Transportation Statistics
- Historical US Data

### Comments:[edit]

- All analysis should be in SI (metric) units if appropriate.
- All research should be properly cited in a standard bibliographic format. All writing should be original, with sparing use of quotes from other sources. Plagiarism will result in an "F".
- A complete history of any technology or mode is the work of a doctoral dissertation, not a term paper. Be as complete and as thorough as possible in the time and space allowed; your instructor understands the constraints you face.
- Start early, last minute work shows. This is a large assignment, pace accordingly.
- Your word processor has spelling and grammar checkers, use them. Attention to detail in writing and presentation is important to persuade the reader you have attended to detail in the analysis. Get a classmate to proofread your work if you have trouble doing so yourself.
- If you have any questions, email or see your instructor, he is happy to help and read and comment on preliminary drafts.
- You are free to talk to classmates about the assignment, but your work must be your own.
- Visit (yes, physically visit) the library. Use multiple reference sources, preferably primary sources, including web and at least 5 non-web sources. Do not use or cite online encyclopedias as primary sources (although they are fine to read and get ideas from). Cite references as appropriate in the text. Cite also the source for your data when you present your data.

Please create a page with your results at Transportation Deployment Casebook.

[note: adapted from Garrison and Hansen UC Berkeley CE 250 Assignment]