12-706 Civil Systems Investment Planning and Pricing/
19-702 Quantitative Methods for Policy Analysis
updated 8/26/2007

This is an overview course designed to introduce students from various backgrounds to the fundamental principles and quantitative methods used in engineering and policy analysis. We will consider economic-based decision making methods, such as decision trees, simulation, sensitivity, benefit-cost and cost-effectiveness analysis. Such methods allow us to determine when decisions should be made by including consideration of relevant economic aspects. The integration of uncertainty into formal methods is a fundamental component of the course. The primary applications in this course will deal with infrastructure systems planning and environmental policy.

This is not intended to be the only course a student would require in quantitative methods, rather it builds a framework for students to build their studies or research. By the end of the course, students will be able to conduct quantitative and qualitative assessments of policy decisions including modeling, visualization, and communication of results.

Required Pre-Requisite Courses

It is assumed that students have had college-level courses on probability and statistics. The Clemen text is a useful refresher of these concepts, not a substitute for them. In addition, engineering economics (capital investment analysis) is a core part of the course. I will have several review sessions on Fridays at the beginning of the semester, but if you have not had this material you should consider at least auditing 12-411 (Engineering Economics). If you have had a course like this already, you might still choose to come to the Friday sessions as a refresher.

Course Audience and Structure

The diversity of the course over time has proven to be a strength. The economics curriculum has mostly theoretical courses, with this being one of the few public project/decision evaluation courses. For graduate students, this course represents a very practical, tool-oriented alternative to some of the other qualitative policy courses.

Textbooks:

• Clemen: "Making Hard Decisions", Clemen and Reilly, Duxbury/Thomson, ISBN 0534365973 (hardback only).
• Sepulveda et al: "Schaum's Outlines of Engineering Economics", McGraw-Hill, ISBN 0070238340 (paperback, $18). Note this is a "study guide" type book on the subject of engineering economics. For those of you who have never learned this material, this guide might be very useful. However, in this case I would really recommend taking 12-411 concurrently.
• I will also put scanned PDF files online from an older Engineering Economics book (written by a former CEE faculty member), for those of you who need more than the study guide type treatment in the Sepulveda text, and an excerpt from the FE Exam Reference Handbook.

Software:

Students will be expected to use Excel-based plugins for this course. Note that the end of many chapters in the Clemen text have detailed tutorials on how to use these programs. We will spend some time in class demonstrating the software tools, but will not necessarily do the same tutorial as the book, or show the same features. We will also spend a few Friday review sessions with software tutorials. Regardless, you should read these parts of the chapters to learn to use the tools - and you might want to bring your laptop to these sessions.

Case studies of relevance to the tools we will develop in the course are widely available in the library on the Internet. An extended bibliography of studies is available. Newspapers such as The Wall Street Journal and The New York Times regularly report on such studies. News journals such as The Economist or ENR (Engineering News Record) are also good sources to read. Planning studies are available in the libraries of funding agencies (e.g. Port Authority of Allegheny County or US Army Corps of Engineers). Professional journals such as the ASCE Journal of Infrastructure Systems are also good sources. We will hand out and review several case studies.

The Syllabus, lecture notes, problem sets and some readings are at http://www.ce.cmu.edu/~hsm/bca2007/. Where possible, all material handed out in class will be posted to the web site. Lecture notes for the entire semester will be available as downloadable PowerPoint presentations. Ideally, slides will be posted in the morning before class (if they are not posted, you might look at previous year lectures by changing the URL to bca2005, etc.). Students should still come to lecture, since the slide files only show overviews of the material being presented and additional instruction will be done, e.g., on the board. I encourage you to download and bring copies of these presentations to class to follow along. If you elect to do this, please help conserve paper by printing in PowerPoint's 'Handouts' option (2, 3, or 6 slides per page) and with duplex (double-sided) printing.

All students should recognize that this class will be taught as a graduate course. For those of you unfamiliar with what this means, here is a short summary:

• All students are expected to read chapters or other handouts before each lecture
• I will lecture on the topic you read, and will lead discussions on the topic
• In lecture I will not generally review the text and may use examples not done in the book
• There may be material/readings assigned that will not be discussed much at all.

In short, this means that while we will follow the textbooks, I will not be simply reteaching the material in each chapter. Coming to lecture will not be a substitute for the reading, and vice versa. I will make more detailed lecture notes/annotations available for certain key concepts to help you study (e.g., welfare economics and decision trees). If you are uncomfortable with this arrangement, you should quickly decide whether to remain registered in the course.

Two primary components of this course will be to build decision models and to improve your skills in solving open-ended problems. Many homework questions will not guide you step-by-step through to a solution - instead you will be presented with an open-ended problem, and be expected to solve it through documentation of your assumptions, modeling, and thought process. You will be graded on both your documentation of assumptions as well as the presentation of your final answer.

The following chart shows the expected distribution of points in the course.

• Grading: Problem Sets - 40%, Case Study Summaries - 10%, Final Exam - 20%, Course Projects - 30%
• There will be approximately 5 case studies done in class where you are expected to read the case and sketch a model solution. For each of these you will need to submit a one page summary of your thoughts and findings.
• Class participation is not an official component of your grade; however, it will help your final grades (e.g. from a high B to a low A).
• Assignments are due in class on the designated date (or in my office for final projects). The deliverable is a paper copy of your work. E-mailing your solutions is not sufficient. Homeworks submitted late will incur a penalty of 10% of total points per day late.
• Collaboration is working together to come to frame a problem and work through a solution, discussing results, and analyzing the process. All members of the group contribute, understand the process (sometimes by being taught by other members of the group), and are prepared to complete a similar problem by themselves afterward. Collaboration is encouraged on homework assignments and in studying for the midterm and final exams. However, for individual assignments, individual submissions are required. While you may have worked with another in solving the problem the work you hand in must be your own.
• Cheating is copying someone else's work and handing it in as your own work. This is unacceptable.
• Plagiarism is using someone else's published work and not giving them credit. Several web sources or the library have guidelines for referencing work from published journals, books, or newspapers, and from websites.
• Cheating and plagiarism will be handled according to university policies, which include the penalty for the assignment (usually a zero grade), and reporting the incident to Student Affairs.
• Regrades are possible within a one week timeframe after the assignments are returned in class. After that time, no regrades will be considered (except for simple addition errors). To submit a regrade, you should attach a sheet of paper detailing your concerns about the score given and argue your point. Regrade requests should be handed to me in class and not put in my mailbox or under my door.

This preliminary course schedule is provided to you in the hopes that it will help you prepare for the first few weeks of class. There will certainly be changes to this schedule as the semester goes on based on the pace of class. The course home page will always have the most up-to-date version, and changes will be announced in class.

The topics, assignments, and suggested readings are given below. I suggest that you read the listed Chapters in column 4 before class to prepare you for lecture. The sample problems will not be collected, but will help guide your thinking for what will be covered in class and on problem sets. I realize that there will be times when you are unable to keep up - however you should do your best not to get too far behind.

Other Topics

The home page gives the proposed schedule, but the course will cover Project Finance (Discounting), Welfare Economics/Benefit-Cost Analysis, Cost-Effectiveness Analysis, Decision Trees, Sensitivity/Uncertainty Analysis, Multi-attribute decisions, and Valuation.

Course Projects

There will be about 3 course projects, each due after roughly 1/3 of the course has been completed. These projects are like take-home group case studies that involve setting up a decision problem, modeling it, and writing a group report summarizing your approach, model, and results. We will rotate group members for these projects. I will distribute more information when the first project is assigned.

For all of the project reports, the accompanying text or visuals should be clear, easy to read, and neat. I suggest you follow (and will use for grading purposes) the CIT Rubric for Writing Technical Reports. However, you do not need to go overboard in formatting the document to look like the work of consultants (i.e., please do not buy special plastic report covers or have them printed at Kinko's on nice paper).

Final Thoughts

Don't let all of the required details above scare you. I have been teaching this course for 10 years - it is my favorite class to teach. I expect we will have a great time together.